Method for generating a blood vessel model
By generating vascular models through level set segmentation and secondary segmentation, the problem of high manpower consumption in existing technologies is solved, achieving efficient and accurate vascular model reconstruction and expanding the application scope of hemodynamic simulation.
Patent Information
- Application Number
- CN202211723122.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-30
- Publication Date
- 2026-05-15
- Estimated Expiration
- 2042-12-30
AI Technical Summary
Existing vascular model reconstruction techniques require a large amount of manual operation and are easily affected by human factors, making them difficult to apply efficiently in clinical settings.
The level set segmentation method is used to segment and reconstruct three-dimensional medical images. It automatically selects the blood vessel branches of interest, extracts the center line and radius, and generates an initial blood vessel branch model. A precise blood vessel model is generated through secondary level set segmentation, avoiding tedious preprocessing operations.
It reduces labor costs, improves model generation speed and accuracy, lowers the technical threshold for hemodynamic simulation, and expands its application scenarios.
Smart Images
Figure CN116129044B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of medical image processing technology, and in particular to a method for generating a blood vessel model. Background Technology
[0002] Hemodynamic simulation technology has broad application prospects in the auxiliary diagnosis and decision-making of cardiovascular and cerebrovascular diseases. However, the high requirements of hemodynamic simulation on vascular models also limit its further application in clinical scenarios. First, the vascular model must be accurate to obtain reliable calculation results. Second, small perforators that do not affect the results must be removed to save computational costs.
[0003] Current techniques for reconstructing vascular models mainly involve segmenting the image using traditional image segmentation algorithms or artificial intelligence algorithms, then reconstructing the surface using the traveling cubes algorithm, and finally performing operations such as trimming, patching, and smoothing on the model. The last step typically consumes a significant amount of manpower, and human interaction can be affected by subjective factors. Summary of the Invention
[0004] Therefore, it is necessary to provide a method for generating a blood vessel model to address the aforementioned technical problems.
[0005] The method for generating a vascular model in this application includes: performing initial level set segmentation on a three-dimensional medical image containing blood vessels to obtain a first binary image sequence containing all vascular lumens;
[0006] Select the branch of interest from the vascular model generated based on the first binary image sequence;
[0007] Extract the centerline and radius along the line of the branch of interest;
[0008] An initial horizontal set of blood vessel branches of interest is generated using each centerline and the radius along the line. Then, a second horizontal set segmentation is performed to obtain a second binary map sequence containing only the blood vessel branches of interest. After reconstruction, a blood vessel model is generated.
[0009] Optionally, the three-dimensional medical image containing blood vessels undergoes initial level set segmentation to obtain a first binary image sequence containing all blood vessel lumens, specifically including:
[0010] The three-dimensional medical image is segmented and reconstructed sequentially using the thresholding method and the traveling cubes method to obtain an initial level set;
[0011] Using the blood vessel walls of the three-dimensional medical image as the zero level set, the first level set segmentation is performed based on the initial level set to obtain a first binary image sequence containing all blood vessel lumens.
[0012] Optionally, selecting the branch of interest includes: receiving a clipping location specified by the user, performing clipping on the vascular model generated based on the first binary image sequence, and obtaining the branch of interest.
[0013] Optionally, the cutting location includes the proximal inlet and distal outlet of the blood vessel branch of interest.
[0014] Optionally, extracting the centerline and radius along the line of the branch of interest includes: obtaining a Vernon map of the region of interest, obtaining the centerline and inscribed sphere along the line of the branch of interest based on the Vernon map, and obtaining the radius along the line based on the inscribed sphere.
[0015] Optionally, an initial level set of blood vessel branches of interest is generated using each centerline and its radius. Then, a second level set segmentation is performed to obtain a second binary map sequence containing only the blood vessel branches of interest. This sequence is then reconstructed to generate a blood vessel model, specifically including:
[0016] The initial model of the branch of the vessel of interest is generated by sweeping along each centerline;
[0017] Obtain the level set based on the initial model of each branch of the vessel of interest;
[0018] The image is segmented using the level set based on the initial blood vessel branch model of interest. The segmentation results are then merged and reconstructed to generate a blood vessel model.
[0019] Optionally, the generation method further includes: taking the intersection of the data of at least one of the first binary image sequence and / or the second binary image sequence with the two-dimensional data of each layer of the three-dimensional medical image, and displaying the intersection marker in the three-dimensional medical image.
[0020] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method for generating a blood vessel model.
[0021] This application also provides an apparatus for generating a blood vessel model, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps:
[0022] The thresholding method and the traveling cubes method are used to segment and reconstruct the three-dimensional medical image sequentially to obtain the initial level set;
[0023] Using the blood vessel walls of the three-dimensional medical image as the zero level set, the first level set segmentation is performed based on the initial level set to obtain the first binary map sequence containing all blood vessel lumens.
[0024] Select the branch of interest from the vascular model generated based on the first binary image sequence;
[0025] Extract the centerline and radius along the line for the branch of interest in the blood vessel;
[0026] The initial model of the branch of the vessel of interest is generated by sweeping along each centerline;
[0027] Obtain the level set based on the initial model of each branch of the vessel of interest;
[0028] The image is segmented using the level set based on the initial blood vessel branch model of interest. The segmentation results are then merged and reconstructed to generate a blood vessel model.
[0029] Optionally, the vascular model generated based on the first binary image sequence is cropped to obtain the branch of interest.
[0030] The method for generating the vascular model in this application has at least the following effects:
[0031] This application avoids manual preprocessing operations such as trimming, hole filling, and smoothing, saving labor costs. When selecting the branch of interest, perforating vessels are automatically deleted, eliminating the need for manual model trimming; an initial set of level images of the branch of interest is generated based on each centerline, resulting in a second binary map sequence for reconstruction, without the need for hole filling and smoothing.
[0032] The method for generating blood vessel models provided in this application avoids cumbersome preprocessing operations and improves the generation speed of the model while ensuring the accuracy of the blood vessel model. Attached Figure Description
[0033] Figure 1 This is a flowchart illustrating a method for generating a blood vessel model in one embodiment of this application;
[0034] Figure 2 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation
[0035] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0036] To solve the above technical problems, please refer to Figure 1 One embodiment of this application provides a method for generating a blood vessel model, including steps S100 to S400. Wherein:
[0037] Step S100: Perform initial level set segmentation on the three-dimensional medical image containing blood vessels to obtain a first binary image sequence containing all blood vessel lumens.
[0038] Step S200: Select the branch of interest from the vascular model generated based on the first binary map sequence.
[0039] Step S300: Extract the centerline and radius along the line of the blood vessel of interest.
[0040] Step S400: Generate an initial horizontal set of blood vessel branches of interest using each centerline and the radius along the line, and then perform secondary horizontal set segmentation to obtain a second binary map sequence containing only the blood vessel branches of interest, and generate a blood vessel model after reconstruction.
[0041] In the embodiments of this application, the initial level set segmentation and the secondary level set segmentation utilize pixel value gradient information at the blood vessel contour in the three-dimensional medical image to re-segment the image, making the zero level set nearly coincide with the blood vessel wall, ultimately obtaining a segmented binary blood vessel image. The model obtained after three-dimensional reconstruction of this image has a smooth surface.
[0042] In this embodiment, the centerline is extracted from the branch of the vessel of interest in the vascular model generated by the first binary image sequence. An initial horizontal set of the branch of interest is then generated based on the centerline, and the vascular model is generated after secondary segmentation of the horizontal set. This also means that this embodiment does not require specific processing of small perforating vessels, as perforating vessels are automatically deleted when generating the initial horizontal set of the branch of interest, avoiding the complex process of manually screening out perforating vessels.
[0043] The vascular model generation method in this embodiment significantly reduces the manpower and complexity required for reconstruction, and the resulting model is less susceptible to human factors, exhibiting high robustness. This embodiment lowers the technical threshold for hemodynamic simulation, broadens its application scenarios, and makes its application in real-time intraoperative scenarios possible.
[0044] The following section provides a combined description of the optional or alternative sub-steps for different steps.
[0045] Step S100 specifically includes steps S110 to S120.
[0046] Step S110: The three-dimensional medical image is segmented and reconstructed sequentially using the thresholding method and the traveling cubes method to obtain an initial level set.
[0047] Step S120: The blood vessel wall of the three-dimensional medical image is used as the zero level set, and the first level set segmentation is performed based on the initial level set to obtain a first binary image sequence containing all blood vessel lumens.
[0048] In this embodiment, the first binary image sequence is obtained based on the threshold-based level set segmentation method, and the second binary image sequence is obtained based on the initial level set segmentation method of the blood vessel branch of interest. Both the threshold-based level set segmentation method and the initial level set segmentation method of the blood vessel branch of interest are iterative calculations performed in a region close to the real blood vessel contour, which can save the time of the algorithm and balance accuracy and efficiency.
[0049] Specifically, a thresholding method is used to perform the first segmentation of the image, and the moving cube algorithm is used to reconstruct the surface of the segmented image to obtain the initial level set for the first level set segmentation process. The threshold can be a fixed gray value or a fixed percentage, such as 15% of the maximum gray value in the image.
[0050] Because the threshold method is very sensitive to the choice of threshold, the model obtained by the threshold method often differs significantly from the real blood vessel boundary due to the combined effects of objective factors such as contrast agent inhomogeneity and partial volume effect, as well as subjective factors such as the user's cognitive bias of the blood vessel boundary.
[0051] In step S120, the initial level set constructed based on the first segmentation (segmentation using the threshold method) is used to iteratively search for the contour with the largest gray value gradient in the image using the level set method. Then, the image is binarized and segmented using this contour as the boundary as the zero level set to obtain the first binary image sequence, thereby making the reconstructed blood vessel model more accurate and repeatable.
[0052] For step S200, select the branch of interest from the vascular model generated from the first binary map sequence;
[0053] Further, selecting the branch of interest includes: receiving a user-specified clipping location, performing clipping on the vascular model generated based on the first binary map sequence to obtain the branch of interest. The clipping location includes the proximal entrance and distal exit of the branch of interest.
[0054] The first binary image sequence obtained in the previous section includes all blood vessels. In this step, the branch of interest is selected through cropping. After receiving the cropping location specified by the user, the proximal and distal ends of the branch of interest are cropped according to the needs of hemodynamic simulation. The blood vessel of interest includes the segment where the lesion is located, as well as adjacent segments and connected major branches, but excludes very small perforating vessels and segments far from the lesion location.
[0055] For step S300, the centerline and radius along the line are extracted for the blood vessel branch of interest.
[0056] As can be seen from the context, the steps requiring user intervention in each embodiment include specifying the cropping location. This simple interactive operation has virtually no impact on the time required to generate the vascular model.
[0057] In step S300, the centerline and radius along the line of the region of interest are extracted, including: obtaining the Vernon map of the region of interest, obtaining the centerline and the inscribed sphere along the line of the blood vessel branch of interest based on the Vernon map, and obtaining the radius along the line based on the inscribed sphere along the line.
[0058] The region of interest typically includes a near-end entrance and multiple far-end exits. This step calculates the Velone diagram from the near-end entrance to each far-end exit.
[0059] Based on the various Vernon diagrams, we obtain the centerline from the near-end entrance to each far-end exit and the corresponding array of maximum inscribed sphere radii, thus obtaining the inscribed spheres along the line and their radii. The maximum inscribed sphere is the inscribed sphere along the centerline with points on the centerline as its centers. The sequence of radii of the maximum inscribed spheres corresponding to all points on the centerline constitutes the array of maximum inscribed sphere radii.
[0060] Step S400 specifically includes steps S410 to S430. Wherein:
[0061] Step S410: Sweep along each centerline to generate an initial model of the branch of the vessel of interest;
[0062] Step S420: Obtain the level set based on the initial model of each branch of the vessel of interest;
[0063] Step S430: The image is segmented twice using the level set based on the initial blood vessel branch model of interest. The segmentation results are merged and reconstructed to generate a blood vessel model.
[0064] Specifically, in steps S410 to S430, sweeping is performed along each centerline to generate initial models of blood vessel branches of interest, each composed of triangular facets. The diameter variation of the initial blood vessel branches of interest can be uniform or follow a non-uniform distribution. For example, the diameter of the vessel at each point on the centerline can be equal to the maximum inscribed sphere radius at that point multiplied by a constant (e.g., 0.9). To improve the efficiency of generating the initial blood vessel branch models of interest, the average, minimum, or median of all elements in the maximum inscribed sphere radius array of the entire blood vessel branch (i.e., the inscribed sphere radius along the line) is used as the sampling step size to sample the centerline.
[0065] Steps S420-S430 involve defining initial level sets of the original 3D image using the surface of each initial branch of interest vessel model, then segmenting the image using the level set method to obtain a sequence of binary images corresponding to each vessel branch—a binary image sequence. The maximum value of all the binary image sequences corresponding to vessels is then processed (i.e., the union of the binary image sequences corresponding to each vessel is taken) to obtain the final binary image sequence. Finally, the moving cube algorithm is used to reconstruct the surface of this binary image sequence, resulting in a vessel model composed of the branches of interest vessels.
[0066] Maximum value processing is performed using the following formula:
[0067] G(x,y)=max(G1(x,y),G2(x,y),...,G n (x, y))
[0068] Where n is the number of branches, Gm(x,y) is the gray value of one of the branches at coordinates (x,y) in the binary image sequence, m is 1 to n, and G(x,y) is the gray value after merging different branches.
[0069] In one embodiment, the method for generating a vascular model, in addition to steps S100 to S400, further includes: taking the intersection of the data of at least one of the first binary image sequence and / or the second binary image sequence with the two-dimensional data of each layer of the three-dimensional medical image, and displaying the intersection marker in the three-dimensional medical image.
[0070] This embodiment displays the reconstructed vascular models at each stage, including those based on thresholding, the first level set segmentation, and the second level set segmentation, in a semi-transparent spatial overlap with the original 3D image. The coordinates of the intersecting pixels between the reconstructed vascular models and each slice of the 3D image are used to locate these intersecting pixels and highlight them, facilitating observation of the proximity between the highlighted edges of the vascular models and the vascular boundaries in the image. Furthermore, the embodiment includes real-time updates of the intersection markers based on the reconstructed vascular models at different stages. For example, during thresholding segmentation, the segmented image and the reconstructed model are updated in real-time by adjusting the threshold.
[0071] It should be understood that, Figure 1 Each step in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.
[0072] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 2 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for generating a blood vessel model. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.
[0073] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:
[0074] Step S100: Perform initial level set segmentation on the three-dimensional medical image containing blood vessels to obtain a first binary image sequence containing all blood vessel lumens.
[0075] Step S200: Select the branch of interest from the vascular model generated based on the first binary map sequence.
[0076] Step S300: Extract the centerline and radius along the line of the blood vessel of interest.
[0077] Step S400: Generate an initial horizontal set of blood vessel branches of interest using each centerline and the radius along the line, and then perform secondary horizontal set segmentation to obtain a second binary map sequence containing only the blood vessel branches of interest, and generate a blood vessel model after reconstruction.
[0078] In one embodiment, a blood vessel model generation apparatus is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the blood vessel model generation method described above. Further details are omitted here. Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0079] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered to be within the scope of this specification. When technical features of different embodiments are embodied in the same drawing, it can be regarded as the drawing also disclosing examples of combinations of the various embodiments involved.
[0080] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.
Claims
1. A method for generating a vascular model, characterized in that, include: The first level set segmentation is performed on the three-dimensional medical image containing blood vessels to obtain the first binary map sequence containing all blood vessel lumens; Select the branch of interest from the vascular model generated based on the first binary image sequence; Extract the centerline and radius along the line of the branch of interest; The initial model of the branch of the vessel of interest is generated by sweeping along each centerline; Obtain the level set based on the initial model of each branch of the vessel of interest; Based on the level set of the initial vessel branch model of interest, the original image is segmented into a second level set to obtain a second binary image sequence that contains only the vessel branch of interest and automatically excludes perforating vessels. The second binary image sequence is surface reconstructed to generate a blood vessel model that does not require manual trimming or patching.
2. The method for generating a vascular model according to claim 1, characterized in that, The three-dimensional medical image containing blood vessels is subjected to initial level set segmentation to obtain a first binary image sequence containing all blood vessel lumens, specifically including: The three-dimensional medical image is segmented and reconstructed sequentially using the thresholding method and the traveling cubes method to obtain an initial level set; Using the blood vessel walls of the three-dimensional medical image as the zero level set, the first level set segmentation is performed based on the initial level set to obtain a first binary image sequence containing all blood vessel lumens.
3. The method for generating a vascular model according to claim 1, characterized in that, The step of selecting the branch of interest includes: receiving a clipping location specified by the user, performing clipping on the vascular model generated based on the first binary image sequence, and obtaining the branch of interest.
4. The method for generating a vascular model according to claim 3, characterized in that, The cutting locations include the proximal inlet and distal outlet of the vessel branch of interest.
5. The method for generating a vascular model according to claim 1, characterized in that, Extracting the centerline and radius along the line of the branch of interest includes: obtaining a Vernon map of the region of interest; obtaining the centerline and inscribed sphere along the line of the branch of interest based on the Vernon map; and obtaining the radius along the line based on the inscribed sphere.
6. The method for generating a vascular model according to claim 1, characterized in that, The generation method further includes: taking the intersection of the data of at least one of the first binary image sequence and / or the second binary image sequence with the two-dimensional data of each layer of the three-dimensional medical image, and displaying the intersection marker in the three-dimensional medical image.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method for generating a vascular model according to any one of claims 1 to 6.
8. An apparatus for generating a blood vessel model, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it performs the following steps: The thresholding method and the traveling cubes method are used to segment and reconstruct the three-dimensional medical image sequentially to obtain the initial level set; Using the blood vessel walls of the three-dimensional medical image as the zero level set, the first level set segmentation is performed based on the initial level set to obtain the first binary map sequence containing all blood vessel lumens. Select the branch of interest from the vascular model generated based on the first binary image sequence; Extract the centerline and radius along the line for the branch of interest in the blood vessel; The initial model of the branch of the vessel of interest is generated by sweeping along each centerline; Obtain the level set based on the initial model of each branch of the vessel of interest; Based on the level set of the initial blood vessel branch model of interest, the original image is segmented into a second level set to obtain a second binary image sequence that contains only the blood vessel branch of interest and automatically excludes perforating vessels.
9. The apparatus for generating a vascular model according to claim 8, characterized in that, The vascular model generated based on the first binary image sequence is pruned to obtain the branch of the vascular interest.