Vascular distortion quantitative method and device based on point cloud aggregation degree index, medium and program product

Abstract

The invention provides a vascular distortion degree quantification method and device based on a point cloud aggregation degree index, a medium and a program product, and relates to the field of intelligent medical treatment. The method comprises the steps of obtaining a geometric model of a to-be-tested blood vessel; extracting a blood vessel three-dimensional center line based on the geometric model; receiving information of a target point on the three-dimensional center line, determining a to-be-measured adjacent range with the target point as the center and the radius being R, and calculating the volume within the radius being R; calculating the local density of the target point based on the target point information, the to-be-measured adjacent range and the volume; determining N nearest neighbor points away from the target point in the adjacent range to be measured; calculating the average distance from the target point to the N nearest neighbor points; calculating an aggregation degree index according to the local density and the average distance; and evaluating the torsion degree of the blood vessel region where the target point is located according to the aggregation degree index. According to the method, the aggregation degree index is introduced, the local density and the geometric features of the adjacent points are comprehensively considered, and the local torsion degree of the blood vessel can be accurately described.

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 quantifying vascular tortuosity based on point cloud aggregation degree index. Background Technology

[0002] Vascular tortuosity refers to the degree to which a blood vessel deviates from a straight line in three-dimensional space, usually expressed by quantifying the tortuous shape of the vessel. It reflects abnormalities in the geometric shape of the blood vessel and may be caused by a variety of factors, such as vascular wall lesions, hemodynamic changes, or congenital developmental abnormalities.

[0003] In existing technologies, the quantitative description of vascular tortuosity mainly focuses on traditional geometric analysis methods (such as the ratio of the length of the spatial curve to the distance from the starting point, the three-dimensional curvature of the spatial curve, and torsion) to describe vascular morphology and tortuosity. For example, the method of the ratio of the length of the spatial curve to the distance from the starting point described by Hoi et al. (2008), the integral method of curvature per unit arc length described by Jakob et al. (1996), and the maximum offset per unit distance described by Wolf et al. (2001) all use the ratio of the length of the spatial curve to the distance from the starting point to reflect the tortuosity of the blood vessel. Although the above methods can reflect the overall degree of tortuosity of the blood vessel, they are insufficient in describing the local details and spatial features of the blood vessel, resulting in the same S / L measurement value for different blood vessels. S represents the spatial arc length of the blood vessel centerline, and L represents the straight-line distance between the starting point and the ending point.

[0004] The above methods typically rely on the geometric features of a single blood vessel point for analysis, lacking a comprehensive consideration of the complex morphology of blood vessels and the geometric features of adjacent points. This makes it impossible to accurately capture the local tortuosity of blood vessels, thus affecting the accurate assessment of blood vessel tortuosity. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention provides a method, apparatus, medium, and program product for quantifying vascular tortuosity based on point cloud aggregation degree index.

[0006] The first aspect of this application discloses a method for quantitatively measuring vascular tortuosity based on point cloud aggregation index, the method comprising:

[0007] S101, Obtain the geometric model of the blood vessel to be tested; Extract the three-dimensional centerline of the blood vessel based on the geometric model;

[0008] S102: Receive target point information located on the three-dimensional centerline, determine the adjacent range to be measured with radius R centered on the target point, and calculate the volume within radius R; calculate the local density of the target point based on the target point information, the adjacent range to be measured, and the volume.

[0009] S103, determine the N nearest neighbor points of the target point within the adjacent range to be tested; calculate the average distance from the target point to the N nearest neighbor points;

[0010] S104, calculate the aggregation index based on the local density and average distance;

[0011] S105, assess the tortuosity of the vascular region where the target point is located based on the aggregation index.

[0012] In some embodiments, the local density is calculated by dividing the number of points within a radius R of the target point by the volume;

[0013] Optionally, the radius R is set according to the global structural scale of the blood vessel to be tested.

[0014] In some embodiments, the method for calculating the average distance includes: In the formula, d(P0, Pi) is the distance between the target point P0 and the nearest neighbor point Pi, and K is the number of nearest neighbors.

[0015] In some embodiments, the clustering index is obtained by dividing the local density by the average distance.

[0016] In some embodiments, in S105, if the aggregation index is higher than a first threshold, the target point is found to be in a densely and centrally distributed region with high distortion.

[0017] In some embodiments, between S101 and S102, the method further includes: receiving a discrete step length and discretizing the three-dimensional centerline into M points, where N is less than M;

[0018] Optionally, N and M are both natural numbers greater than 1.

[0019] In some embodiments, between S101 and S102, the method further includes preprocessing the three-dimensional centerline with one or more of the following: B-spline curve, least squares method, Gaussian filtering, and connectivity analysis.

[0020] Optionally, the three-dimensional centerline is the main trunk centerline after removing irrelevant branches;

[0021] Optionally, the geometric model is constructed based on medical image data.

[0022] A second aspect of this application discloses a computer device, the device comprising: a memory and a processor; the memory being used to store a computer program; and the processor executing the computer program to implement the steps of the above-described method.

[0023] 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.

[0024] 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.

[0025] The shortcomings of existing technologies and the beneficial effects of this application:

[0026] 1. Consider only local changes

[0027] Limitations of existing technologies: Traditional methods assess vascular tortuosity by calculating the ratio of the length of a spatial curve to the distance from its starting point. This primarily focuses on changes in the overall morphology and fails to capture local morphological changes at different locations along the curve. Especially in complex vascular morphologies, details such as local bending and twisting cannot be accurately reflected by this ratio, thus failing to provide the specific degree of tortuosity at each vascular point.

[0028] Improvements of this patent: This patent introduces an aggregation index, which comprehensively considers local density and the geometric features of adjacent points, enabling a precise description of the local tortuosity of blood vessels. The aggregation index provides a more detailed quantitative description at each vascular point, capturing changes in local morphology, especially in complex vascular regions, such as tortuous or intersecting areas, providing a more accurate quantification of tortuosity.

[0029] 2. Inability to demonstrate the spatial distribution of local distortion.

[0030] Limitations of existing technologies: Existing methods typically only calculate the local curvature or torsion at a single point, without considering the influence of the geometric features of adjacent points on the tortuosity of that point. In complex vascular morphologies, multiple intersecting or tortuous vascular segments may lead to an inaccurate description of the overall tortuosity, and local changes may not be effectively captured.

[0031] Improvements of this patent: This patent introduces a clustering index, combining the geometric features and average distance of adjacent points in adjacent regions. This not only captures the overall morphological change trend but also provides a detailed description of the local tortuosity of blood vessels. By combining local density, the geometric relationship of neighboring points, and global information, this method can more comprehensively assess the tortuosity of blood vessels, especially in complex morphological regions, reflecting the spatial distribution of local tortuosity.

[0032] 3. Lack of consideration for the internal and external relationships of vascular structures.

[0033] Limitations of existing technologies: While existing methods consider hemodynamics (such as blood flow pressure and velocity), they primarily rely on complex physical models and typically neglect the geometric relationships between local vascular structures and adjacent vessels. Therefore, these methods cannot provide a comprehensive quantitative assessment of the overall tortuosity of blood vessels and lack microscopic details revealing changes in vascular morphology.

[0034] Improvements of this patent: This patent, through an aggregation index, comprehensively considers local density, the geometric features of adjacent points, and the morphological structure of blood vessels. This not only overcomes the shortcomings of traditional methods in their lack of sensitivity to local changes but also takes into account the geometric relationships between adjacent blood vessel points. Therefore, the comprehensive index of this patent can provide a more comprehensive and accurate quantitative assessment of the overall tortuosity of blood vessels, especially offering higher accuracy in complex structural regions (such as intersections and bends). Attached Figure Description

[0035] 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.

[0036] Figure 1 This is a schematic diagram of the method flow provided in the first aspect of the present invention;

[0037] Figure 2 This is a schematic diagram of a blood vessel tortuosity quantification system based on point cloud aggregation index provided in the second aspect of the present invention;

[0038] Figure 3 This is a schematic diagram of a computer device provided in an embodiment of the present invention;

[0039] Figure 4 This is a schematic diagram of the architecture of an exemplary computing device provided in an embodiment of the present invention;

[0040] Figure 5 This is a schematic diagram of the storage medium provided in an embodiment of the present invention;

[0041] Figure 6 This is a geometric calculation method for representing the degree of blood vessel tortuosity in the prior art provided by the embodiments of the present invention; wherein, Figure 6 A represents Method 1: the ratio of the length of the spatial curve to the distance from the starting point, S represents the spatial arc length of the blood vessel centerline, and L represents the straight-line distance between the starting point and the ending point. Figure 6 Method B represents the second method: the unit arc length curvature integral method, where k is the curvature on the curve and S is the total arc length, used as a normalization factor. Figure 6C represents the third method: the maximum offset per unit distance method, where δ represents the maximum offset distance from a point on the curve to the line connecting the start and end points (i.e., the straight line L0), and L0 represents the straight-line distance between the start and end points;

[0042] 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 in Figure 1.

[0043] Figure 8 This refers to the aggregation degree provided in the embodiments of the present invention when the radius R and the number of nearest neighbors K are 5, 10, 20, and 30, respectively; wherein, Figure 8 A represents the aggregation degree when K is 5. Figure 8 B represents the aggregation degree when K is 10. Figure 8 C represents the aggregation degree when K is 20. Figure 8 D represents the aggregation degree when K is 30. Detailed Implementation

[0044] 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.

[0045] 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.

[0046] 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.

[0047] Figure 1 This is a schematic flowchart of a method for quantifying vascular tortuosity based on point cloud aggregation index provided by an embodiment of the present invention. Specifically, the method includes the following steps: S101, obtaining the geometric model of the blood vessel to be tested; extracting the three-dimensional centerline of the blood vessel based on the geometric model;

[0048] In some embodiments, the blood vessels to be tested include, but are not limited to, the following: aorta, superior vena cava, and inferior vena cava.

[0049] In some embodiments, between S101 and S102, the method further includes: receiving a discrete step length and discretizing the three-dimensional centerline into M points, where N is less than M;

[0050] Optionally, N and M are both natural numbers greater than 1.

[0051] It should be noted that the blood vessel to be tested 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.

[0052] 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.

[0053] In some embodiments, between S101 and S102, the method further includes preprocessing the three-dimensional centerline with one or more of the following: B-spline curve, least squares method, Gaussian filtering, and connectivity analysis.

[0054] Optionally, the three-dimensional centerline is the main trunk centerline after removing irrelevant branches;

[0055] Optionally, the geometric model is constructed based on medical image data.

[0056] S102: Receive target point information located on the three-dimensional centerline, determine the adjacent range to be measured with radius R centered on the target point, and calculate the volume within radius R; calculate the local density of the target point based on the target point information, the adjacent range to be measured, and the volume.

[0057] In some embodiments, the local density is calculated by dividing the number of target points within a radius R by the volume; the radius R is set according to the global structural scale of the blood vessel to be tested.

[0058] S103, determine the N nearest neighbor points of the target point within the adjacent range to be tested; calculate the average distance from the target point to the N nearest neighbor points;

[0059] In some embodiments, the method for calculating the average distance includes: In the formula, d(P0, Pi) is the distance between the target point P0 and the nearest neighbor point Pi, and K is the number of nearest neighbors.

[0060] S104, calculate the clustering index based on the local density and average distance; in some embodiments, the clustering index is obtained by dividing the local density by the average distance.

[0061] S105, assess the tortuosity of the vascular region where the target point is located based on the aggregation index.

[0062] In some embodiments, in S105, if the aggregation index is higher than a first threshold, the target point is found to be in a densely and centrally distributed region with high distortion.

[0063] In some embodiments, the first threshold is obtained by training with training set samples. It can be a specific threshold or an interval range. The specific form is not specifically limited in this embodiment.

[0064] Figure 3 This is a schematic diagram of a computer device provided in an embodiment of the present invention, such as... Figure 3 As 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.

[0065] 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.

[0066] 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.

[0067] 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.

[0068] 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.

[0069] 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.

[0070] In some embodiments, this embodiment also discloses a quantitative system for vascular tortuosity based on point cloud aggregation index, such as... Figure 2 As shown, the system includes:

[0071] The model acquisition and centerline extraction module 201 acquires the geometric model of the blood vessel to be tested; and extracts the three-dimensional centerline of the blood vessel based on the geometric model.

[0072] The local density calculation module 202 receives target point information located on the three-dimensional centerline, determines the adjacent range to be measured with radius R centered on the target point, and calculates the volume within the radius R; and calculates the local density of the target point based on the target point information, the adjacent range to be measured, and the volume.

[0073] The average distance calculation module 203 determines the N nearest neighbor points of the target point within the adjacent range to be measured; and calculates the average distance from the target point to the N nearest neighbor points.

[0074] The clustering index calculation module 204 calculates the clustering index based on the local density and average distance.

[0075] The vascular tortuosity assessment module 205 assesses the tortuosity of the vascular region where the target point is located based on the aggregation index. Specific Implementation

[0076] Step 1: Acquisition of medical image data and construction of vascular geometric models

[0077] 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.

[0078] 1. Data Loading and Preprocessing

[0079] 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.

[0080] 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.

[0081] 2. Extract isosurfaces

[0082] 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.

[0083] 3. Generate point cloud data

[0084] 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, describing the surface shape of the region of interest. Taking the aortic trunk as an example, this step involves dividing the centerline of the aortic trunk extracted in the previous step into a series of discrete points and identifying their 3D coordinates, preparing for subsequent torsion assessment.

[0085] 4. Blood vessel segmentation and extraction

[0086] 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.

[0087] 5. 3D Reconstruction and Model Generation

[0088] 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.

[0089] Step 2: Extract the centerline of the three-dimensional blood vessel

[0090] 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.

[0091] 1. Skeletonization and Centerline Extraction

[0092] 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.

[0093] 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.

[0094] 2. Centerline smoothing and optimization

[0095] 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.

[0096] 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.

[0097] 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.

[0098] 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.

[0099] 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.

[0100] 5. Verification and Adjustment

[0101] The accuracy of the extracted centerline is verified through visual inspection or comparison with the actual morphology of blood vessels. The consistency between the extracted centerline and the actual morphology of the blood vessel can be checked by aligning and comparing it with the vessel's edge. Based on the verification results, the segmentation threshold, distance transformation parameters, or refinement algorithm parameters are adjusted to optimize the centerline extraction effect.

[0102] By employing image preprocessing, skeletonization, and 3D centerline extraction methods, the 3D centerline of blood vessels can be extracted from CT scan data. The centerline is not only fundamental to vascular morphology analysis but also a crucial step in further studies such as vascular tortuosity and hemodynamics. Using appropriate algorithms and tools, accurate vascular centerlines can be obtained, providing a reliable data foundation for subsequent geometric feature calculations.

[0103] Step 3: Determine the discrete length and uniformly discrete the vessel centerline.

[0104] The extracted vessel centerline is discretized into multiple points with a step size of 1 mm (e.g., a 200 mm centerline is discretized into 200 points). This step helps to accurately calculate the local geometric features of the vessel curve.

[0105] 1. Calculate local density

[0106] Define a clustering index to measure the local density of each point:

[0107]

[0108] Where R represents the adjacent range (e.g., 20 mm), N(P0, R) represents the number of points within the radius R of the target point P0, and VR is the volume within that radius (the observable range). The radius R is set according to the scale of the global vascular structure.

[0109] 2. Calculate the average distance between adjacent points.

[0110] Calculate the average distance from each point to its 20 nearest neighbors:

[0111]

[0112] Where d(P0, P) i () is point P0 and point P i The distance between them, K is the number of nearest neighbors (20 in this case).

[0113] 3. Calculate the clustering index

[0114] To quantify clustering, a clustering index is defined, which combines local density, average neighbor distance, and global point distribution. Taking into account both the neighborhood size and local density of each point, a comprehensive clustering index C(P0) is defined:

[0115]

[0116] This index can reflect both the density of points and the distribution of neighboring points. A higher clustering index indicates that point P0 is located in a relatively dense and concentrated area. Figure 8 The aggregation degree is defined as follows: for a radius R, the number of nearest neighbors K is 5, 10, 20, and 30.

[0117] Key technical points or key technical advantages of this patent

[0118] The following are the specific effects brought about by this technology:

[0119] 1. Improved quantitative accuracy of vascular tortuosity: This patent innovatively combines the aggregation index with local density and geometric features of adjacent points to more accurately quantify the local tortuosity of blood vessels. Compared with traditional methods, the aggregation index provides higher accuracy in areas with complex vascular morphologies (such as vascular intersections and bends), avoiding the shortcomings of traditional methods in quantitative description of these complex morphologies.

[0120] 2. Enhanced Comprehensive Analysis of Local and Global Information: This patent not only considers the overall morphological changes of blood vessels but also comprehensively describes the quantitative degree of vascular tortuosity by integrating local and global geometric features. The aggregation index can consider the geometric features of adjacent points, breaking through the limitations of traditional methods that rely solely on the geometric features of a single blood vessel point. Especially in complex vascular structures, it can accurately capture the specific degree of tortuosity of each blood vessel point.

[0121] 3. Support for Early Diagnosis and Personalized Treatment: This patent provides more precise quantitative data for the early diagnosis and treatment of vascular diseases. By accurately measuring vascular tortuosity, it helps doctors identify potential vascular disease risks and supports personalized treatment and surgical planning. The aggregation index can effectively predict the occurrence of vascular diseases, providing patients with more targeted diagnostic and treatment options.

[0122] 4. Advancing vascular imaging technology: This technology provides new insights into vascular morphology research, promoting the application of medical imaging in the diagnosis of vascular diseases. Using this technology, doctors can obtain more detailed and precise vascular morphology analysis, thereby improving the efficiency and accuracy of vascular disease diagnosis and treatment.

[0123] Specifically, such as Figure 6 and Figure 7 As shown, Figure 6 Existing geometric calculation methods for representing the degree of vascular tortuosity (Hoiet et al., 2008; Jakob et al., 1996; Wolf et al., 2001) can only consider local variations, failing to reflect the spatial distribution of local tortuosity and lacking consideration of the internal and external relationships of the vascular structure; while Figure 7 middle, 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.

[0124] The following are the key technologies brought about by this technology:

[0125] 1. Innovative Application of Aggregation Index: The aggregation index proposed in this invention is a comprehensive evaluation method based on local density, average distance between adjacent points, and global point distribution. It can accurately quantify vascular tortuosity, and is particularly suitable for the analysis of complex vascular morphologies. The core value of this innovation lies in its ability to provide detailed local and global vascular tortuosity information, overcoming the shortcomings of traditional methods.

[0126] 2. Comprehensive consideration of local and global vascular morphology: By introducing an analysis method that combines local and global perspectives, this invention can more accurately describe changes in vascular morphology, especially in complex regions (such as intersections and tortuosities). The influence of the geometric features of adjacent vascular points on vascular tortuosity is fully considered, thereby improving the accuracy of quantitative description.

[0127] 3. Precise calculation based on three-dimensional vascular geometric model: This invention provides an accurate mathematical basis for the analysis of vascular tortuosity by using a three-dimensional vascular geometric model, combined with image data, centerline extraction, smoothing optimization and other algorithms, ensuring high precision and high reliability of the quantification results.

[0128] 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.

[0129] 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.

[0130] 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.

[0131] 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.

[0132] 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.

[0133] 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.

[0134] 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 quantitatively determining vascular tortuosity based on point cloud aggregation index, characterized in that, The method includes: S101, Obtain the geometric model of the blood vessel to be tested; Extract the three-dimensional centerline of the blood vessel based on the geometric model; S102: Receive target point information located on the three-dimensional centerline, determine the adjacent range to be measured with radius R centered on the target point, and calculate the volume within radius R; calculate the local density of the target point based on the target point information, the adjacent range to be measured, and the volume. S103, determine the N nearest neighbor points of the target point within the adjacent range to be tested; calculate the average distance from the target point to the N nearest neighbor points; S104, calculate the aggregation index based on the local density and average distance; S105, assess the tortuosity of the vascular region where the target point is located based on the aggregation index.

2. The method for quantitatively determining vascular tortuosity based on point cloud aggregation index according to claim 1, characterized in that, The local density is calculated by dividing the number of points within the radius R of the target point by the volume. Optionally, the radius R is set according to the global structural scale of the blood vessel to be tested.

3. The method for quantitatively determining vascular tortuosity based on point cloud aggregation index according to claim 1, characterized in that, The method for calculating the average distance includes: In the formula, d(P0, Pi) is the distance between the target point P0 and the nearest neighbor point Pi, and K is the number of nearest neighbors.

4. The method for quantitatively determining vascular tortuosity based on point cloud aggregation index according to claim 1, characterized in that, The clustering index is obtained by dividing the local density by the average distance.

5. The method for quantitatively determining vascular tortuosity based on point cloud aggregation index according to claim 1, characterized in that, In step S105, if the aggregation index is higher than the first threshold, the target point is found to be in a densely distributed region with high distortion.

6. The method for quantitatively determining vascular tortuosity based on point cloud aggregation index according to claim 1, characterized in that, Between S101 and S102, the method further includes: receiving discrete step lengths and discretizing the three-dimensional centerline into M points, where N is less than M; Optionally, N and M are both natural numbers greater than 1.

7. The method for quantitatively determining vascular tortuosity based on point cloud aggregation index according to claim 1, characterized in that, Between S101 and S102, the method further includes preprocessing the three-dimensional centerline by any one or more of B-spline curves, least squares method, Gaussian filtering and connectivity analysis. Optionally, the three-dimensional centerline is the main trunk centerline after removing irrelevant branches; Optionally, the geometric model is constructed based on medical image data.

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