Method, device, medium and program product for describing blood vessel morphology by integrating local and global information

Abstract

The embodiment of the invention provides a method, equipment, a medium and a program product for describing a blood vessel form by integrating local and global information, and relates to the field of intelligent medical treatment. The method comprises: acquiring a blood vessel center line of a target blood vessel region; the blood vessel center line is a geometric center track of a blood vessel lumen; dispersing the center line of the blood vessel into N points according to a first step length to obtain N discrete points; determining a to-be-measured discrete point and an adjacent range, and generating K discrete points in the adjacent range by taking the to-be-measured discrete point as a center; for each point Pi in the M discrete points, calculating a discrete point Pi-1 and a discrete point Pi + 1 adjacent to the point Pi, and calculating a local included angle of vectors at two ends; the average value of the local included angles is calculated, and the direction change rate of the to-be-measured discrete point in the space field is obtained; and evaluating the form of the blood vessel region where the to-be-detected discrete point is located according to the direction change rate.

Description

Technical Field

[0001] This invention relates to the field of intelligent healthcare, and more specifically, to a method, device, medium, and program product for describing vascular morphology by integrating local and global information. Background Technology

[0002] In the field of quantitative description of vascular morphology, existing technologies mainly rely on vascular morphology analysis methods based on three-dimensional curvature and torsion or vascular morphology analysis methods based on dynamic vascular characteristics to assess vascular morphology.

[0003] For example, invention patent application number 202010919099.0 describes a method and system for assessing aortic tortuosity based on differential geometry. This method utilizes differential geometry theory to describe the degree of vascular curvature or torsion through local curvature, thereby inferring the degree of vascular tortuosity. By introducing the mathematical concepts of three-dimensional curvature and torsion, this method characterizes the local bending and twisting of blood vessels in space, overcoming the shortcomings of traditional methods for assessing aortic tortuosity that neglect local details. Other studies analyze morphological changes through the dynamic behavior of blood vessels, using hemodynamic models combined with the geometric characteristics of the vessels to calculate the degree of vascular tortuosity. For example, the deformation and tortuosity of blood vessels are derived based on parameters such as blood flow pressure and velocity.

[0004] However, none of the above methods fully consider the geometric relationships between adjacent points, ignoring spatial features such as point cloud distribution and direction change rate in the neighborhood. In complex areas such as vascular intersections and severe torsion, single-point features are easily interfered with by noise or local anomalies, leading to inaccurate judgment of local vascular morphology. Summary of the Invention

[0005] The present invention aims to at least solve one of the technical problems existing in the prior art. To this end, the present invention provides a method, apparatus, medium, and program product for describing vascular morphology by integrating local and global information.

[0006] The first aspect of this application discloses a method for describing vascular morphology by integrating local and global information, the method comprising:

[0007] S101, Obtain the centerline of the target vascular region; the centerline of the vascular region is the geometric center trajectory of the vascular lumen;

[0008] S102, Discretize the blood vessel centerline into N points according to the first step length to obtain N discrete points;

[0009] S103, determine the discrete point to be measured and the adjacent range, and generate K discrete points in the adjacent range with the discrete point to be measured as the center;

[0010] S104, for each point P among M discrete points iCalculate the relationship with point P i Adjacent discrete points P i-1 and discrete point P i+1 Calculate the local angle between the two vectors;

[0011] S105, calculate the average value of the local included angles to obtain the rate of change of direction of the discrete point to be measured in the spatial domain;

[0012] S106 assesses the morphology of the vascular region where the discrete point is located based on the rate of change of direction. The more the direction changes, the more distorted the region, averaging the angles between adjacent points.

[0013] In some embodiments, the local included angle is calculated as follows: In the formula, This reflects the centerline at P i The degree of curvature at the point ranges from 0° to 180°.

[0014] In some embodiments, the rate of change of direction is calculated as follows: In the formula, P0 is the discrete point to be measured, and N(P0, R) represents the number of points P0 within the radius R, where the radius R is the adjacent range.

[0015] Optionally, the adjacent range R is set according to the global structural scale of the target blood vessel.

[0016] In some embodiments, the discretization method includes uniform discretization and non-uniform discretization;

[0017] Optionally, M is less than or equal to N, and K is less than or equal to M; N, M, and K are all natural numbers greater than 1.

[0018] In some embodiments, the method further includes:

[0019] Based on the discrete point P0 to be measured, calculate its local density in the adjacent range R;

[0020] Based on the discrete point P0 to be measured, determine the K nearest discrete points to P0; calculate the average distance from the discrete point P0 to the K nearest discrete points.

[0021] The overall aggregation degree is calculated based on the rate of change of direction, local density, and average distance.

[0022] The morphology of the vascular region where the discrete point to be tested is located is assessed based on the comprehensive aggregation degree.

[0023] Optionally, the local density can be calculated by: calculating the volume of the discrete point to be measured within the adjacent range R; and calculating the ratio of K to the volume.

[0024] In some embodiments, the method for calculating the overall aggregation degree is as follows:

[0025] The ratio of local density to average distance is calculated as the local index;

[0026] The overall aggregation degree is calculated based on the local index and the rate of change of direction; specifically: In the formula, Local indices representing local density and average distance; It is a weighting coefficient used to balance the importance of local density, average distance, and rate of change of direction.

[0027] In some embodiments, the vascular centerline is the main trunk centerline after removing irrelevant branches.

[0028] A second aspect of this application discloses a computer device, comprising: a memory and a processor; the memory is used to store a computer program; the processor executes the computer program to implement the steps of the above-described method.

[0029] A third aspect of this application discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0030] The fourth aspect of this application discloses a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0031] This application has the following beneficial effects: 1. This application innovatively discloses a method based on the rate of change of direction in the spatial domain of a discrete point within a target vascular region, introducing local angle features and global angle features, effectively capturing the local morphological changes of the curve at different positions, considering the angle features of adjacent points, and capturing the changing trend of the blood vessel from the overall morphology.

[0032] 2. This application innovatively integrates local and global information to describe vascular morphology. At the local scale, it calculates local density and local angles as core features to accurately reflect the spatial distribution characteristics of blood vessels within a local area, such as bending, twisting, or abrupt changes in direction. It also calculates the average distance between adjacent points and the average angle within the spatial neighborhood, characterizing the overall continuity and directional changes of the vascular path. This multi-scale clustering analysis is a typical local-global integrated modeling method. This method is suitable for analyzing geometric distortions in complex vascular regions and exhibits good sensitivity, stability, and interpretability. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a schematic diagram of the method flow provided in the first aspect of the present invention;

[0035] Figure 2 This is a schematic diagram of a system for describing vascular morphology by integrating local and global information, as provided in the second aspect of the present invention.

[0036] Figure 3 This is a schematic diagram of a computer device provided in an embodiment of the present invention;

[0037] Figure 4 This is a schematic diagram of the architecture of an exemplary computing device provided in an embodiment of the present invention;

[0038] Figure 5 This is a schematic diagram of the storage medium provided in an embodiment of the present invention;

[0039] Figure 6 This invention provides a vascular centerline and a three-dimensional line graph of discrete points that discretize the vascular centerline into multiple points; wherein, Figure 6 A represents the central line of the blood vessel. Figure 6 B is a three-dimensional line graph of discrete points using uniform sampling at 0.1 mm. Figure 6 C is a three-dimensional line graph of discrete points using 1 mm uniform sampling. Figure 6 D is a three-dimensional line graph of discrete points using 5 mm uniform sampling;

[0040] Figure 7 This is a schematic diagram illustrating the differences in the aggregation degree of vascular centerlines with different spatial morphologies at the same evaluation point, provided by an embodiment of the present invention; wherein, Figure 7 A represents a curve with a stable spatial shape. Figure 7 B represents the spatial morphological distortion curve. Two evaluation points at the same location and with the same neighborhood radius R are selected. Three-dimensional curvature and torsion methods cannot distinguish between the two, while the aggregation index proposed in this patent can accurately reflect the higher local distortion characteristics of region B. Detailed Implementation

[0041] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0042] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Figure 1 This is a schematic flowchart of a method for describing blood vessel morphology by integrating local and global information according to an embodiment of the present invention. Specifically, the method includes the following steps:

[0045] S101, Obtain the centerline of the target vascular region; the centerline of the vascular region is the geometric center trajectory of the vascular lumen;

[0046] In some embodiments, the vascular centerline is the main trunk centerline after removing irrelevant branches.

[0047] It should be noted that the target vascular region is obtained based on the geometric model of the blood vessel. The geometric model is constructed based on medical image data. The medical image data can come from 3D or 4D medical image data of one or more imaging modes of the patient. In this embodiment, computed tomography (CTA) medical image data is used; of course, it can also be any other type of medical imaging mode, such as computed tomography (CTA), magnetic resonance imaging, angiography, ultrasound imaging and other medical imaging modes.

[0048] It should be noted that, taking the aortic trunk as an example, the aortic trunk is the portion of the aorta extending from its inlet to the iliac branches. CTA image data of the aorta includes not only the aorta itself but also surrounding muscle tissue and non-aortic vessels. The purpose of this step is to separate the aorta from other background areas in the image of the human aorta, extracting the aorta—that is, obtaining the region of interest—and performing 3D reconstruction of this region of interest. Next, the centerline of the trunk of the reconstructed anatomical model is extracted to prepare for subsequent distortion analysis. The specific steps for establishing the aortic anatomical model and extracting the trunk centerline are described in detail below.

[0049] In some embodiments, the target vascular region used herein is derived from a “subject” or “test subject” or “sample to be tested,” which refers to any animal (e.g., a mammal), including but not limited to humans, non-human primates, rodents, etc., who will become the recipient of a particular treatment. Generally, the terms “subject” and “patient” are used interchangeably herein when referring to a human subject. Preferably, the subject is a human.

[0050] S102, Discretize the blood vessel centerline into N points according to the first step length to obtain N discrete points;

[0051] In some embodiments, the discretization method includes uniform discretization and non-uniform discretization;

[0052] It should be noted that, taking the aortic trunk as an example, this step involves segmenting the centerline of the aortic trunk extracted in the previous steps into a series of discrete points and identifying their three-dimensional coordinates, in preparation for subsequent torsion assessment. The specific steps for discretizing the aortic centerline and identifying the three-dimensional coordinates of each discrete point will be described in detail below.

[0053] S103, determine the discrete point to be measured and the adjacent range, and generate K discrete points in the adjacent range with the discrete point to be measured as the center;

[0054] Optionally, M is less than or equal to N, and K is less than or equal to M; N, M, and K are all natural numbers greater than 1.

[0055] S104, for each point P among M discrete points i Calculate the relationship with point P i Adjacent discrete points P i-1 and discrete point P i+1 Calculate the local angle between the two vectors;

[0056] In some embodiments, the local included angle is calculated as follows: In the formula, This reflects the centerline at P iThe degree of curvature at the point ranges from 0° to 180°.

[0057] S105, calculate the average value of the local included angles to obtain the rate of change of direction of the discrete point to be measured in the spatial domain;

[0058] In some embodiments, the rate of change of direction is calculated as follows: In the formula, P0 is the discrete point to be measured, and N(P0, R) represents the number of points P0 within the radius R, where the radius R is the adjacent range.

[0059] Optionally, the adjacent range R is set according to the global structural scale of the target blood vessel.

[0060] S106, assess the morphology of the vascular region where the discrete point to be tested is located based on the rate of change of direction.

[0061] In some embodiments, the method further includes:

[0062] Based on the discrete point P0 to be measured, calculate its local density in the adjacent range R;

[0063] Based on the discrete point P0 to be measured, determine the K nearest discrete points to P0; calculate the average distance from the discrete point P0 to the K nearest discrete points.

[0064] The overall aggregation degree is calculated based on the rate of change of direction, local density, and average distance.

[0065] The morphology of the vascular region where the discrete point to be tested is located is assessed based on the comprehensive aggregation degree.

[0066] Optionally, the local density can be calculated by: calculating the volume of the discrete point to be measured within the adjacent range R; and calculating the ratio of K to the volume.

[0067] In some embodiments, the method for calculating the overall aggregation degree is as follows:

[0068] The ratio of local density to average distance is calculated as the local index;

[0069] The overall aggregation degree is calculated based on the local index and the rate of change of direction; specifically: In the formula, Local indices representing local density and average distance; It is a weighting coefficient used to balance the importance of local density, average distance, and rate of change of direction.

[0070] Figure 3 This is a schematic diagram of a computer device provided in an embodiment of the present invention, such as... Figure 3As shown, the device 2000 may include: one or more processors 2010 and one or more memories 2020; wherein the memories store computer-readable code that, when run by the one or more processors, can perform the methods described above.

[0071] The processor in this embodiment can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, operations, and logic block diagrams disclosed in this embodiment. The general-purpose processor can be a microprocessor or any conventional processor, and can be based on an x86 or ARM architecture.

[0072] In general, the various exemplary embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, firmware, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software that can be executed by a controller, microprocessor, or other computing device. When aspects of embodiments of this disclosure are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.

[0073] For example, the method or apparatus according to embodiments of this disclosure can also be used by means of Figure 4 The architecture of the computing device 3000 shown is used for implementation. For example... Figure 4 As shown, the computing device 3000 may include a bus 3010, one or more CPUs 3020, a read-only memory (ROM) 3030, a random access memory (RAM) 3040, a communication port 3050 connected to a network, an input / output component 3060, a hard disk 3070, etc. The storage devices in the computing device 3000, such as the ROM 3030 or the hard disk 3070, may store various data or files used for processing and / or communication of the methods provided in this disclosure, as well as program instructions executed by the CPU. The computing device 3000 may also include a user interface 3080. Of course, Figure 4 The architecture shown is merely exemplary and can be omitted as needed when implementing different devices. Figure 4 One or more components in the computing device shown.

[0074] This invention also includes a computer-readable storage medium, such as... Figure 5The diagram illustrates a storage medium 4000 provided in an embodiment of the present invention. The computer storage medium 4020 stores computer-readable instructions 4010. When the computer-readable instructions 4010 are executed by a processor, the method described above according to embodiments of the present disclosure can be performed. The computer-readable storage medium in the embodiments of the present disclosure may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Synchronous Link Dynamic Random Access Memory (SLDRAM), and Direct Memory Bus Random Access Memory (DR RAM). It should be noted that the memory used in the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0075] This disclosure also provides a computer program product or system, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0076] In some embodiments, this embodiment also discloses a system that integrates local and global information to describe vascular morphology, such as... Figure 2 As shown, the system includes:

[0077] The vascular centerline acquisition module 201 is used or configured to acquire the vascular centerline of the target vascular region; the vascular centerline is the geometric center trajectory of the vascular lumen;

[0078] The discrete point generation module 202 is used or configured to discretize the blood vessel centerline into N points according to the first step length, thereby obtaining N discrete points;

[0079] The adjacent range determination module 203 is used or configured to determine the discrete point to be measured and the adjacent range, and generate K discrete points within the adjacent range with the discrete point to be measured as the center;

[0080] Local angle calculation module 204, used or configured for each point P among M discrete points. i Calculate the relationship with point Pi Adjacent discrete points P i-1 and discrete point P i+1 Calculate the local angle between the two vectors;

[0081] The orientation change rate calculation module 205 is used or configured to calculate the average value of the local included angle to obtain the orientation change rate of the discrete point to be measured in the spatial domain;

[0082] The first vascular morphology assessment module 206 is used or configured to assess the morphology of the vascular region where the discrete point to be tested is located based on the rate of change of direction.

[0083] In some embodiments, the system further includes:

[0084] The local density calculation module is used or configured to calculate the local density of the discrete point P0 to be measured within the adjacent range R.

[0085] The average distance calculation module is used or configured to determine the K nearest discrete points to the discrete point P0 to be measured, and calculate the average distance from the discrete point P0 to the K nearest discrete points.

[0086] The comprehensive clustering calculation module is used or configured to calculate the comprehensive clustering based on the rate of change of direction, local density, and average distance.

[0087] The second vascular morphology assessment module is used or configured to assess the morphology of the vascular region where the discrete point to be tested is located based on the comprehensive aggregation degree. Specific Implementation

[0089] Step 1: Acquisition of medical image data and construction of vascular geometric models

[0090] The DICOM file format is widely used to store medical images, such as CT and MRI scans. It also contains image-related metadata (e.g., patient information, scanning equipment information, image resolution, etc.). In VTK, we first load the DICOM file as a `vtkImageData` object using a DICOM data reader. This object represents three-dimensional volumetric data, where each voxel (the corresponding point of a pixel in the image in three-dimensional space) has a grayscale or density value, typically used to represent different types of human tissue or lesion areas.

[0091] 1. Data Loading and Preprocessing

[0092] The first step in extracting the centerline of the vascular model is to use VTK to load and preprocess the medical image data. VTK supports various medical image data formats, including DICOM and NIfTI. First, the image data is loaded into VTK's vtkImageData structure, which stores voxel information of the 3D image, suitable for subsequent image processing and analysis.

[0093] The loaded image may contain noise and unnecessary details, so it needs to be smoothed after loading the image data. VTK provides Gaussian filtering (vtkImageGaussianSmooth) and other image smoothing filters to remove noise from the image and smooth image edges, ensuring more accurate subsequent vessel segmentation and centerline extraction.

[0094] 2. Extract isosurfaces

[0095] In DICOM data, each voxel contains a grayscale value, while the density of certain regions in medical images (such as tumors and organs) typically falls within a fixed range. Therefore, we need to extract these specific regions from the 3D volumetric data. VTK provides algorithms such as `vtkContourFilter` and `vtkMarchingCubes`, which can extract regions of interest based on a set threshold. These regions are represented by isosurfaces, which are surfaces formed by all points in 3D space with the same grayscale value. For example, by setting a threshold, all regions with grayscale values ​​greater than that value can be extracted, resulting in a 3D surface model, commonly used to represent organ boundaries, lesion areas, etc.

[0096] 3. Generate point cloud data

[0097] When a surface mesh is generated using an isosurface extraction algorithm (such as `vtkContourFilter`), each mesh face consists of multiple triangles, and the vertices of these triangles represent the points of interest. These points define the 3D geometry of the extracted region. By extracting the vertex coordinates from the mesh, we obtain a point cloud of discrete points. Each point corresponds to a location in 3D space, and these points describe the surface shape of the region of interest.

[0098] 4. Blood vessel segmentation and extraction

[0099] Blood vessels are distinguished from the background based on their grayscale (or density) values. In vascular images, blood vessels typically have high density or grayscale values. By setting an appropriate threshold, areas with grayscale values ​​greater than a certain threshold are marked as vascular regions. VTK provides the vtkImageThreshold algorithm to perform this thresholding segmentation, which can effectively separate vascular regions from the surrounding tissue and background. During image segmentation, especially for images with complex backgrounds or that are relatively blurry, edge detection (such as the Sobel algorithm) can help identify the contours of blood vessels. Region growing algorithms can further refine the segmentation of vascular regions by expanding from a seed point based on pixel similarity. VTK's vtkImageSobel3D and vtkImageConnectivityFilter can help extract the edges or connected regions of blood vessels, further refining the segmentation results.

[0100] 5. 3D Reconstruction and Model Generation

[0101] After image segmentation, 3D reconstruction of the vascular region is fundamental for centerline extraction. Using the vtkMarchingCubes algorithm, VTK can extract isosurfaces from voxel data to generate a surface model of the blood vessel. This process converts the segmented vascular region into a 3D mesh model, allowing subsequent centerline extraction to be performed in 3D space. This algorithm provides the geometric basis for centerline extraction by constructing a surface mesh model of the blood vessel. In the generated 3D vascular model, the surface of the vessel wall is represented as a triangular mesh, and centerline extraction is further performed within this surface.

[0102] Step 2: Extract the centerline of the three-dimensional blood vessel

[0103] Image processing algorithms are used to extract the centerline of blood vessels. The centerline is a key element describing the structure of blood vessels and provides the basis for calculating the geometric features of blood vessels, such as tortuosity. Extracting the centerline of three-dimensional blood vessels is a crucial step in vascular morphological analysis, as it forms the basis for subsequent geometric feature calculations.

[0104] 1. Skeletonization and Centerline Extraction

[0105] Skeletonization is the process of extracting the central axis of a blood vessel from its surface model. Its purpose is to simplify the three-dimensional surface of the vessel into a thin line representing its geometric center, thus forming the centerline. By progressively "eroding" the vessel surface, a simplified central axis is ultimately retained; this is the vessel's central axis. Skeletonization incorporates distance transformation techniques to calculate the shortest distance from each point to the vessel's edge, thereby deriving the vessel's geometric center and aiding in the extraction of the accurate centerline. This process preserves the vessel's topology but removes its width, leaving only the "skeleton," which is helpful for subsequent morphological analysis and tortuosity measurements.

[0106] For 3D CT images, the first step is to perform a 3D distance transformation on the vascular region, calculating the distance from each vascular voxel to the nearest non-vascular voxel. This step ensures that we can obtain accurate geometric information of the vascular structure. Based on this, a 3D skeletonization algorithm is used to simplify the vascular region to a centerline. Commonly used skeletonization algorithms include: Thinning algorithms, which iteratively remove edge pixels from the vascular region to obtain the thinned centerline of the vessel; distance transformation methods, which obtain the maximum distance point from each voxel to the background through distance transformation and use it as the centerline point of the vessel; and graph-based centerline extraction methods, which construct a graph structure and use shortest path algorithms (such as Dijkstra's algorithm) or minimum spanning tree (MST) methods to extract the optimal path from the vascular region as the centerline of the vessel. These algorithms make the extraction of the vascular centerline more accurate, especially in complex vascular structures, effectively extracting the accurate vascular centerline from 3D data, supporting subsequent vascular morphology analysis, disease diagnosis, and surgical planning.

[0107] 2. Centerline smoothing and optimization

[0108] During the extraction of vascular centerlines, the results are often not perfectly smooth, but contain noise and irregularities, which can affect the accuracy of subsequent analysis. Therefore, denoising and smoothing are crucial steps to ensure the accuracy of vascular centerline extraction. To this end, this study employs a series of advanced mathematical and image processing methods to optimize the geometry of vascular centerlines and effectively reduce noise interference.

[0109] B-splines are used to construct a set of piecewise polynomial curves, ensuring the curves remain continuous and smooth across each segment, thus guaranteeing the smoothness of the vessel centerline within local regions. In practical applications, B-splines can finely adjust the smoothness of the curve by controlling the distribution and weights of nodes, avoiding over-smoothing or excessively stiff curves. Furthermore, the interpolation capability of B-splines allows the vessel centerline to accurately fit the original data points, removing unnecessary fluctuations while preserving the vessel's morphological characteristics, ensuring the preservation of the vessel's topological and geometric features.

[0110] The smoothed data were fine-tuned using the least squares method. Least squares minimizes the error at each data point during the fitting process, thus reducing the deviation between the vessel centerline and the actual vessel structure. This method is particularly effective for handling complex vessel structures, reducing noise while ensuring the accuracy of the fitting results, making the vessel path more closely resemble the actual anatomy.

[0111] A Gaussian filtering-based smoothing method was introduced. Gaussian smoothing works by weighting the data points along the centerline, where the weights are determined by a Gaussian function, with data points closer to the center receiving a larger weight and those farther away receiving a smaller weight. This weighted averaging mechanism effectively reduces local fluctuations along the vessel centerline and removes irregularities caused by noise or edge effects. The advantage of Gaussian filtering is that its smoothing effect does not rely on simple point-to-point connections but considers neighborhood information of the data points, thus enabling a more comprehensive handling of complex vascular structures.

[0112] Detecting and removing points with low connectivity through connectivity analysis is another crucial step in ensuring a smooth and coherent centerline. Typically, we use the vtkPolyDataConnectivity Filter to identify and remove isolated points in blood vessels, ensuring a clear and intact central axis. After thresholding small branches or isolated points, we further apply morphological erosion operations (such as vtkImageErosion or vtkPolyDataFilter) to remove excessively small discrete branches, guaranteeing the coherence and accuracy of the centerline.

[0113] 5. Verification and Adjustment

[0114] Centerline extraction is not only fundamental to vascular morphology analysis but also a crucial step in further studies such as vascular tortuosity and hemodynamics. By using appropriate algorithms and tools, accurate vascular centerlines can be obtained, providing a reliable data foundation for subsequent geometric feature calculations. The accuracy of the extracted centerline is verified through visualization or comparison with the actual vascular morphology. Common methods include aligning the extracted centerline with the vessel edge to determine if it accurately reflects the vessel's geometric orientation and branching structure. Based on the verification results, the segmentation threshold, distance transformation parameters, or refinement algorithm parameters are adjusted to optimize the centerline extraction effect.

[0115] Step 3: Uniformly Discretize the Centerline of Blood Vessels

[0116] The extracted vascular centerline is discretized into multiple points (i.e., N discrete points, such as a 200 mm long centerline discretized into 200 points, where N is 200) with a step size of 1 mm. Figure 6 Image A shows the vascular centerline extracted from a 3D image, clearly illustrating the structural outline of the aorta and its major branches. To further evaluate the impact of discretization precision on the accuracy of the centerline, uniform sampling at different intervals was performed, generating 0.1 mm (... Figure 6 B), 1 mm ( Figure 6 C) and 5 mm ( Figure 6 D) is a three-dimensional line graph of discrete points.

[0117] This uniform discretization operation ensures the consistency of the overall point cloud distribution along the centerline, providing a standardized global-scale basis for the accurate calculation of subsequent local density, proximity features, and angle variations. This step helps to accurately calculate the local geometric features of the vascular curve while maintaining the consistency of the overall structure.

[0118] 1. Calculate local density

[0119] For each blood vessel point P0, i.e., the discrete point to be measured, its local density within the radius R is calculated and defined as:

[0120]

[0121] Where R is the adjacent range (e.g., 20mm), N(P0, R) represents the number of points P0 within the radius R, and V R It is the volume (observable range) within this radius. The radius R is set according to the scale of the global vascular structure.

[0122] 2. Calculate the average distance between adjacent points.

[0123] Calculate the average distance from each point P0 to its 20 nearest neighbors:

[0124]

[0125] Where d(P0, P) i () is point P0 and point P i The distance between them is K, and K is the number of nearest neighbors (20 here, but not limited to 20 in practice).

[0126] 3. Calculate local angles

[0127] For the discretized centerline, for each point Calculate its local included angle ,definition , and The angle between the two vectors:

[0128]

[0129] in, Reflects the center line at The degree of curvature at the point ranges from 0° to 180°.

[0130] 4. Average angle within the computational space domain

[0131] For each point Within its space domain Calculate the local angles between all points within radius R. average :

[0132]

[0133] Where N(P0, R) represents the number of points P0 falls within the radius R;

[0134] In practice, local and average angles can be calculated separately to obtain the rate of change of direction of the discrete point in the spatial domain, which is used to assess the morphology of the vascular region where the discrete point is located. When averaging adjacent angles, the greater the change in direction, the more distorted the structure.

[0135] 5. Calculate the clustering index.

[0136] To accurately quantify the local tortuosity of vascular structures, this patent proposes a comprehensive aggregation degree, which comprehensively considers multiple geometric factors, including the local density of points, the average distance between adjacent points, the rate of change of direction, and global reference information on the overall morphology of the blood vessel. This comprehensive aggregation degree is defined as follows: :

[0137]

[0138]

[0139] in, This represents a clustering index based on density and average distance. These are weighting coefficients used to balance the importance of density, distance, and angular features; This is the corrected comprehensive aggregation index. The level of this aggregation index directly reflects the complexity and directional change trend of vascular structures in space. When a region experiences vascular tortuosity, twisting, intersection, or abrupt changes in path, the local point cloud becomes denser, the distance between neighboring points is shorter, and the directional change is more drastic, thus increasing the aggregation index. Significantly increased. Furthermore, the clustering index integrates geometric features at multiple levels. At the local scale, by calculating local density and local angles as core features, it accurately reflects the spatial distribution characteristics of blood vessels within a local area, such as bending, twisting, or abrupt changes in direction. It also calculates the average distance between adjacent points and the average angle within the spatial neighborhood, characterizing the overall continuity and directional changes of the vascular path. This multi-scale clustering analysis is a typical local-global integrated modeling method. This method is suitable for analyzing geometric distortions in complex vascular regions and exhibits good sensitivity, stability, and interpretability.

[0140] Specific effects are as follows Figure 7 As shown, Figure 7 A and Figure 7 B is a schematic diagram of the centerlines of two blood vessels with different spatial morphologies. Two evaluation points and the same neighborhood radius R are selected at the same location (the blue point on the centerline is the target point). If existing three-dimensional curvature and torsion calculation methods are used, since the calculation only depends on the derivative at the blue point, the calculation results will be exactly the same, making it difficult to distinguish the essential differences in their spatial structures. However, from... Figure 7 It can be clearly seen from the middle, Figure 7 B exhibits more complex curvature variations and point cloud density clustering within its neighborhood, clearly indicating that its degree of distortion is higher than that of point cloud density clustering. Figure 7 A. This patent introduces an aggregation index, Figure 7 The aggregation value calculated by B is significantly higher than that calculated by B. Figure 7 A, taking into account the influence of the geometric features of adjacent points, can more realistically reflect the differences in local spatial morphology and effectively support the quantitative assessment of vascular tortuosity.

[0141] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0142] In general, the various exemplary embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, firmware, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software that can be executed by a controller, microprocessor, or other computing device. When aspects of embodiments of this disclosure are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.

[0143] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0144] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0145] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0146] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0147] The exemplary embodiments of this disclosure described in detail above are merely illustrative and not restrictive. Those skilled in the art will understand that various modifications and combinations can be made to these embodiments or their features without departing from the principles and spirit of this disclosure, and such modifications should fall within the scope of this disclosure.

Claims

1. A method for describing vascular morphology by integrating local and global information, characterized in that the method... include: S101, Obtain the center line of the target blood vessel region; the center line of the blood vessel is the geometric center trajectory of the blood vessel lumen; S102, the blood vessel centerline is discretized into N points according to the first step length, resulting in N discrete points; S103, determine the discrete point to be measured and the adjacent range, and generate K discrete points in the adjacent range with the discrete point to be measured as the center; S104, for each point P among M discrete points i Calculate the relationship with point P i Adjacent discrete points P i-1 and discrete point P i+1 Calculate the local angle between the two vectors; S105, calculate the average value of the local included angles to obtain the rate of change of the direction of the discrete point to be measured in the spatial domain; S106, Evaluate the morphology of the vascular region where the discrete point to be tested is located based on the directional change rate.

2. The method for describing vascular morphology by integrating local and global information according to claim 1, characterized in that, The calculation method for the local included angle is as follows: In the formula, This reflects the centerline at P i The degree of curvature at the point ranges from 0° to 180°.

3. The method for describing vascular morphology by integrating local and global information according to claim 2, characterized in that, The method for calculating the rate of change of direction is as follows: In the formula, P0 is the discrete point to be measured, and N(P0, R) represents the number of points P0 falls within the radius R, where the radius R is the adjacent range.

4. The method for describing vascular morphology by integrating local and global information according to claim 1, characterized in that, The discretization methods include uniform discretization and non-uniform discretization; Optionally, M is less than or equal to N, and K is less than or equal to M; N, M, and K are all natural numbers greater than 1.

5. The method for describing vascular morphology by integrating local and global information according to claim 1, characterized in that, The method further includes: Based on the discrete point P0 to be measured, calculate its local density within the adjacent range R; Based on the discrete point P0 to be measured, determine the K nearest discrete points to P0; calculate the average distance from the discrete point P0 to the K nearest discrete points. The overall aggregation degree is calculated based on the directional change rate, local density, and average distance. The morphology of the vascular region where the discrete point to be tested is located is evaluated based on the comprehensive aggregation degree. Optionally, the method for calculating the local density is as follows: calculate the volume of the discrete point to be measured within the adjacent range R; calculate the ratio of K to the volume.

6. The method for describing vascular morphology by integrating local and global information according to claim 5, characterized in that, The method for calculating the overall aggregation degree is as follows: The ratio of the local density to the average distance is calculated as the local exponent; The overall aggregation degree is calculated based on the local index and the rate of change of direction; specifically: In the formula, Local indices representing local density and average distance; It is a weighting coefficient used to balance the importance of local density, average distance, and rate of change of direction.

7. The method for describing vascular morphology by integrating local and global information according to claim 1, characterized in that, The vascular centerline is the centerline of the main trunk after removing irrelevant branches.

8. A computer device, characterized in that, The device includes: a memory and a processor; the memory is used to store a computer program; the processor executes the computer program to implement the steps of the method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method described in any one of claims 1-7.

Images