Carotid artery image-oriented image processing and simulation analysis method

Abstract

The invention discloses an image processing and simulation analysis method for a carotid artery image, which comprises the following steps: firstly, acquiring a CTA image, observing and selecting, then importing the CTA image, adjusting the image contrast, generating a three-dimensional blood vessel model, extracting a carotid artery focus, and carrying out overall and local surface optimization processing on the model. The method comprises the steps that firstly, a model is built, then local surface cutting naming and grid division are conducted on the model, finally parameters are set, CFD hemodynamic simulation is conducted, a simulation result is subjected to visualization processing, and preoperative and postoperative data are compared and analyzed. According to the method, multi-modal imaging data of brain CT perfusion imaging, brain CT angiography and the like of a patient suffering from carotid artery stenosis can be collected to be used for reconstructing a three-dimensional model of the carotid artery stenosis, the carotid artery hemodynamic characteristics of the patient and the individual characteristics of the patient are analyzed in an auxiliary mode based on computational fluid mechanics, operation mode selection is optimized, and the three-dimensional model of the carotid artery stenosis is reconstructed. Cerebral ischemia reperfusion injury is avoided, clinical prognosis of a patient is improved, and life quality of the patient is improved.

Description

Technical Field

[0001] The present invention belongs to the cross - technical field of medical imaging technology and computational fluid dynamics (CFD), and particularly relates to a method for processing and simulation analysis of carotid artery images Background Technique

[0002] Stroke is the leading cause of death in the Chinese population and is a serious neurological disease threatening human life and health. Among them, ischemic stroke accounts for 80%, with the characteristics of high disability rate, high recurrence rate and high mortality rate. One of its main causes is carotid atherosclerotic lesions. As a key blood vessel connecting the brain and the heart, the hemodynamic state of the carotid artery directly affects the blood perfusion of the brain. Once the carotid artery becomes stenotic or blocked, it will lead to insufficient blood supply to the brain, thus triggering ischemic stroke. In the elderly population over 65 years old, the prevalence of carotid artery stenosis is as high as 10%, and about 20 - 30% of carotid artery stenosis patients will experience ischemic stroke events every year, causing a significant public health burden. Approximately 50% of carotid artery stenosis patients have reduced cerebral blood flow and cerebrovascular reactivity due to vascular stenosis, and are prone to vascular - related cognitive impairment caused by cerebral hypoperfusion. Carotid artery stenosis has a slow onset, and its clinical and imaging manifestations are hidden. Long - term existence will seriously affect human health and quality of life. Timely detection and effective treatment of carotid artery lesions are crucial for preventing and controlling stroke. In recent years, the continuous progress of medical imaging technology has provided more accurate means for the diagnosis of carotid artery lesions. However, relying solely on imaging examinations often makes it difficult to comprehensively reflect the overall hemodynamic status of the carotid artery, which also limits the doctor's in - depth analysis of the condition and the formulation of personalized treatment plans

[0003] Currently, the annual incidence of ischemic stroke in China is approximately 2.5 million to 3 million, which has become the leading cause of death and disability among Chinese nationals. Among them, carotid artery stenosis is an important independent risk factor for ischemic stroke. Approximately 50% of carotid artery stenosis patients are prone to vascular - related cognitive impairment caused by cerebral hypoperfusion, seriously affecting human health and quality of life. For the recanalization of moderate - to - severe carotid artery stenosis, there are currently two main surgical methods: carotid endarterectomy (CEA) and carotid artery stenting (CAS). The effects of these two surgical methods on the hemodynamic changes inside and outside the skull before and after surgery, the occurrence mechanism of peri - operative complications, the establishment method of collateral circulation inside and outside the skull, and the long - term prognosis are currently unclear. Therefore, studying the peri - operative risk assessment and clinical efficacy of different surgical methods is of great significance for the prognosis of patients

[0004] Finite element numerical simulation is an important tool for biomechanics research, especially suitable for the hemodynamic research of the carotid artery. Currently, there have been some numerical simulation studies on carotid artery blood flow at home and abroad, but there are relatively few literatures on finite element numerical simulation of the carotid artery for individual data. Therefore, the carotid artery model reconstructed based on CTA images provides a theoretical basis for in-depth understanding of hemodynamic mechanisms and their clinical treatments. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides an image processing and simulation analysis method for carotid artery images, providing a clear and systematic process to help researchers and clinicians understand the entire process of carotid artery three-dimensional model reconstruction and hemodynamic simulation.

[0006] The technical solution adopted by the present invention is as follows: an image processing and simulation analysis method for carotid artery images, and the specific steps are as follows:

[0007] S1. Acquisition and processing of CTA images: Import CTA images into a medical image viewing software for observation and select a suitable CTA image file.

[0008] First, obtain CTA images, that is, inject a contrast agent through a vein and let it reach the blood vessel area to be imaged with the blood. X-rays pass through the patient's body and interact with the contrast agent in the blood vessels to form an image of the blood vessels. Then, find the image with high blood vessel contrast and many slice layers after adding the contrast agent as a file, and record its file name and number.

[0009] Among them, the obtained CTA images, namely CTA image files, are discontinuous layered files in.dcm format formed by CTA technology, including: cerebral CT perfusion imaging and cerebral CT angiography of patients with carotid artery stenosis.

[0010] Then, import the CTA image files for individual observation, find the image with high blood vessel contrast and many slice layers after adding the contrast agent as a file, record its file name and number, and use this type of CTA image file as the original file.

[0011] S2. Three-dimensional model reconstruction: Import the original CTA image selected in step S1, adjust the image contrast, select the case with the largest contrast, extract the blood vessel part, generate a three-dimensional model and perform a first cut to intercept the edge part of the carotid artery.

[0012] S3. Model optimization: Extract the carotid artery lesions and perform overall and local surface optimization processing on the carotid artery lesion model.

[0013] S4. Mesh generation: Perform local surface cutting and naming and mesh division on the model processed in step S3, and then check and correct the surface.

[0014] S5, CFD hemodynamic simulation, importing the grid file, setting the corresponding parameters, and performing simulation;

[0015] S6. Based on the simulation results obtained in step S5, data analysis and visualization are performed, that is, the simulation results are visualized and the pre-operative and post-operative data are compared and analyzed.

[0016] Furthermore, the step S2 is specifically as follows:

[0017] S21, import the original CTA image file and adjust the image contrast;

[0018] Import the original file selected in step S1 into the medical image processing software, adjust the grayscale specification to "Soft tissue scale", enhance the contrast of the medical image in the software, and obtain the required carotid artery blood vessels.

[0019] S22, based on step S21, finding the area where the carotid artery is located, marking it by threshold comparison, and extracting a three-dimensional model of the blood vessel;

[0020] After adjusting the contrast of the image, the blood vessel part represented by the highlighted area is extracted from the two-dimensional image and converted into a three-dimensional model by threshold segmentation. The lower limit of the threshold is adjusted in the image, and the selected content is the brightness between the lower limit of the threshold and the upper limit of the interval. After the threshold range is determined, a mask with operation is formed. The mask is calculated to construct the overall three-dimensional model, and the three sections are compared to see if there is any difference with the three-dimensional model. If there is a difference, the threshold range is redefined and reconstructed.

[0021] In the process of threshold cutting, check the "Keep Largest" option and the "Fill Holes" option. Through "Keep Largest", the largest objects or structures are screened and retained. Through "Fill Holes", the holes or gaps in the 3D model are filled and smoothed and merged according to the surrounding geometric shapes and data structures.

[0022] S23, based on step S22, cutting the extracted three-dimensional blood vessel model once, finding the required edge of the carotid artery, and intercepting it;

[0023] The cutting operation is still performed in the medical image processing software. Use the "Orthogonal to Screen" function in the 3D tool to create a section perpendicular to the screen view on the 3D model. At the same time, the segmented shape is displayed in other colors, and the model before cutting is retained. During the operation, choose to continue cutting or re-cut according to the effect after segmentation.

[0024] Furthermore, the specific steps of step S3 are as follows:

[0025] S31. For the carotid artery edge intercepted in step S23, intercept it again, cut and remove the disordered entities similar to noise generated due to image imaging reasons, and retain the model of the lesion part.

[0026] S32. Based on step S31, smooth the surface of the model of the lesion part as a whole and export it in stl format.

[0027] For the voids, tomographic defects, geometric noise, sharp edges or irregular shapes on the model surface caused by fewer file slices, perform overall smoothing on the entire separately intercepted lesion part.

[0028] By adjusting the contrast iteration times and the smoothing factor parameters, adjust the effect of the overall smoothing process. Under the condition of a suitable smoothing factor, perform smoothing iterations successively to obtain the effect after multiple smoothing process iterations.

[0029] Among them, the condition of a suitable smoothing factor means that each time under this smoothing factor parameter, each smoothing operation iteration will cause an image, but will not significantly change the model.

[0030] S33. Import the carotid artery lesion model after surface smoothing in step S32 into the software of the three-dimensional analysis model and perform local defect smoothing.

[0031] For defects such as sharp edges or irregular shapes that still exist after the overall smoothing process, perform local smoothing operations on the defective part, and perform local smoothing and trimming without affecting other model areas.

[0032] S34. Detect whether there are defects in the model processed in step S33. If there are defects, continue to repair them. If there are no defects, enter step S4.

[0033] Among them, the detection parameters include: Full analysis, Inverted normals, Bad edges, Near badedges, Planar holes, Shells, Overlapping triangles, Intersecting triangles.

[0034] Furthermore, the specific steps of step S4 are as follows:

[0035] S41. Perform end face resection on the repaired model and name the cutting surface and the carotid artery surface.

[0036] Based on Step S3, after the entire surface and local smoothing processes are completed, the original cross-section has deformed. Then, a new cut is made perpendicular to the blood vessel, and the newly generated cross-section is used to define the surface for input and output boundary conditions, and it is named accordingly.

[0037] Among them, for the naming of the boundaries, the cross-section of the common carotid artery where blood flows in is set as in, and the cross-sections of the internal carotid artery and external carotid artery where blood flows out are set as out1 and out2 respectively, and the blood vessel wall is set as wall.

[0038] S42. Based on Step S41, the surface mesh and volume mesh of the model are divided under appropriate parameters;

[0039] For the surface mesh, operations for generating or modifying triangular meshes are performed in "Target triangle edge length", and the item to be modified is the desired size of the triangular edge length. At the same time, parameters are set, including: "Preserve surface contours", "Maximum triangle edge length", "Influence area", "Growth rate".

[0040] Similarly, the parameters set in the volume mesh are the same as those in the surface mesh.

[0041] S43. The model after mesh division in Step S42 is defect-detected again and modified until no defects exist;

[0042] S44. The model with defect-free surface and volume meshes is exported in the format of.mesh.

[0043] Furthermore, Step S5 is specifically as follows:

[0044] S51. The mesh file obtained in Step S4 is imported into the computational fluid dynamics software, and the relevant parameters of the model size, surface, and volume meshes are checked, and the simulation model related to fluid mechanics is selected;

[0045] The k-epsilon model is used as the simulation model, and the parameters of the blood are set, including: blood density, blood viscosity.

[0046] S52. Boundary conditions are defined for the end faces and surfaces named in Step S4;

[0047] In the three-dimensional model reconstruction stage, only four boundaries appear in the model with good cutting and smoothing effects, that is, the four surface positions named in Step S41. The input type of the common carotid artery is selected as velocity input, the output types of the internal carotid artery and external carotid artery are selected as pressure output, and the blood flow velocity parameters detected by the actual device are input.

[0048] S53. Add cross-sections with the narrow part and its front and back as key positions to detect data;

[0049] For the entire inflow and outflow, detect the velocity and compare various parameters at the stenosis before and after the operation. Then create planes parallel to the plane where the xy-axis is located, with positions in front of the carotid artery stenosis, at the carotid artery stenosis, and behind the carotid artery stenosis, and name them "before", "narrow", and "after" in sequence to reflect the differences in the same parameters at different positions contrastively.

[0050] S54. According to the preset accuracy requirements of the experimental data, select appropriate residual values for the simulation of the simulation results;

[0051] S55. Based on Step S54, export the simulation results in the formats of.cas and.dat files.

[0052] Further, the specific steps of Step S6 are as follows:

[0053] S61. Check the imported.cas and.dat files and screen various parameter attempts;

[0054] S62. Select the exported parts and exported parameters;

[0055] S63. Export the parameters of the two simulation results before and after the operation in the format of a 3D picture comparison;

[0056] The selected parameters include: streamline diagram of the flow velocity, wall shear stress, and blood vessel pressure. Adjust the data to be exported according to actual needs, and by default, fully characterize and output single-item data for the best result feedback. If multiple data overlays are required, adjust the parameter ratio according to the actual situation.

[0057] S64. Export the detailed conditions at each position in the form of a scatter plot and export the corresponding result data set;

[0058] The data is saved as a.txt file in numerical form, and select the screenshot button on the side of the interface to save it in the form of a picture and adjust according to actual needs.

[0059] For the pressure parameters in each position direction applied to other calculations, select to save them in the.txt format.

[0060] The parameters derived with the X-axis as the coordinate axis include: "Static Pressure", "Pressure Coefficient", "Dynamic Pressure", "Absolute Pressure", "Total Pressure". The parameters derived with the Y-axis as the coordinate axis include: "Static Pressure", "Wall Whear Stress", "Turbulent Kinetic Energy".

[0061] Advantages of the present invention: The method of the present invention first obtains CTA images and performs observation and selection, then imports CTA images, adjusts the image contrast, generates a three-dimensional vascular model, extracts carotid artery lesions, performs overall and local surface optimization processing on the model, then performs local surface cutting and naming and meshing on the model, and finally sets parameters, performs CFD hemodynamic simulation, and visualizes the simulation results for comparative analysis of pre-operative and post-operative data. The method of the present invention can collect multi-modal imaging data such as cerebral CT perfusion imaging and cerebral CT angiography of patients with carotid artery stenosis, be used to reconstruct the three-dimensional model of carotid artery stenosis, assist in analyzing the carotid artery hemodynamic characteristics and individual characteristics of patients based on computational fluid dynamics, assist doctors in making more scientific clinical decisions, optimize the selection of surgical methods, avoid cerebral ischemia-reperfusion injury, improve the clinical prognosis of patients, improve the quality of life of patients, and at the same time provide a clear and systematic process to help understand the entire process of carotid artery three-dimensional model reconstruction and hemodynamic simulation. Brief Description of the Drawings

[0062] Figure 1 It is a flowchart of a method for image processing and simulation analysis of carotid artery images according to the present invention.

[0063] Figure 2 It is a schematic diagram of high contrast of blood vessels after adding contrast agent in an embodiment of the present invention.

[0064] Figure 3 It is a schematic diagram of comparison before and after gray-scale processing in an embodiment of the present invention.

[0065] Figure 4 It is a schematic diagram of three-dimensional reconstruction of blood vessels in an embodiment of the present invention.

[0066] Figure 5 It is a schematic diagram of comparison of blood vessel cutting effects in an embodiment of the present invention.

[0067] Figure 6 It is a schematic diagram of multiple surface iterations in an embodiment of the present invention.

[0068] Figure 7Schematic diagram of local processing and defect repair in the embodiments of the present invention.

[0069] Figure 8 Schematic diagram of the setting of the grid and the parameters of the volume grid in the embodiments of the present invention.

[0070] Figure 9 Schematic diagram of error detection and defect repair in the embodiments of the present invention.

[0071] Figure 10 Schematic diagram of the selection of boundary condition types in the embodiments of the present invention.

[0072] Figure 11 Schematic diagram of the key reference positions in the embodiments of the present invention.

[0073] Figure 12 Comparison chart of streamline parameters of result visualization processing speed in the embodiments of the present invention.

[0074] Figure 13 Comparison chart of wall shear force parameters of result visualization processing in the embodiments of the present invention.

[0075] Figure 14 Comparison chart of blood vessel pressure parameters of result visualization processing in the embodiments of the present invention.

[0076] Figure 15 Scatter diagram of speed positions of result visualization in the embodiments of the present invention. Detailed implementation manners

[0077] The method of the present invention will be further described below in conjunction with the drawings and embodiments.

[0078] As Figure 1 shown, the flowchart of an image processing and simulation analysis method for carotid artery images of the present invention is as follows:

[0079] S1. Acquisition and processing of CTA (CT angiography) images. Import CTA images in the RadiAnt DICOM Viewer software (medical image viewing software) for observation and select suitable CTA image files;

[0080] First, obtain CTA images. That is, inject the contrast agent intravenously and let it reach the blood vessel area to be imaged with the blood. The X-ray passes through the patient's body and interacts with the contrast agent in the blood vessels to form an image of the blood vessels. When the contrast agent is not added, compared with bones, other organs and tissues do not have strong distinguishability. Therefore, find the image with high blood vessel contrast and many slice layers after adding the contrast agent as the file, and record its file name and number for subsequent processing. The image with the contrast agent added is as Figure 2 shown, and the position of the carotid artery is marked by the red circle in the figure.

[0081] Among them, the obtained CTA image, that is, the CTA image file, is a discontinuous layered file in.dcm format formed by CTA technology, including: cerebral CT perfusion imaging and cerebral CT angiography of patients with carotid artery stenosis.

[0082] Then, import the CTA image file for individual observation, find the image with high blood vessel contrast and many slice layers after adding the contrast agent as the file, record its file name and number, and use this type of CTA image file as the original file.

[0083] S2. Three-dimensional model reconstruction: Import the original CTA image file selected in step S1, adjust the image contrast, select the case with the largest contrast, extract the blood vessel part, generate a three-dimensional model and perform a first cut to intercept the carotid artery edge part;

[0084] S3. Model optimization: Extract the carotid artery lesions and perform overall and local surface optimization processing on the carotid artery lesion model;

[0085] S4. Mesh generation: Perform local surface cutting and naming and mesh division on the model processed in step S3, and then check and correct the surface;

[0086] S5. CFD (Computational Fluid Dynamics) hemodynamic simulation: Import the mesh file, set the corresponding parameters, and perform the simulation;

[0087] S6. Based on the simulation results obtained in step S5, perform data analysis and visualization, that is, visualize the simulation results and compare and analyze the data before and after the operation.

[0088] In this embodiment, the specific steps of step S2 are as follows:

[0089] S21. Import the original CTA image file and adjust the image contrast;

[0090] Import the original file selected in step S1 into Mimics software (medical image processing software), adjust the gray scale specification, enhance the contrast of the medical image in the software, and obtain the required carotid artery.

[0091] As Figure 3 shown, comparing before and after gray scale processing, by changing the gray scale specification to "Softtissue scale" in the contrast adjustment interface, the entire position of the carotid artery with the contrast agent added is clearer than before, with a greater difference from the surrounding area, making it more convenient to distinguish through threshold division, thereby obtaining the required carotid artery;

[0092] S22. Based on step S21, find the area where the carotid artery is located, mark it through threshold comparison, and extract the three-dimensional model of the blood vessel;

[0093] After adjusting the contrast of the picture, the blood vessel part represented by the highlighted area is extracted from the two-dimensional picture through threshold segmentation and converted into a three-dimensional model. Adjust the lower limit of the threshold in the picture, and the selected content is the brightness between the lower limit of the threshold and the upper limit of the interval. After the threshold range is determined, a mask with operations is formed. Calculate the mask to construct the overall three-dimensional model, and compare whether there are differences between the three profiles and the three-dimensional model. If there are differences, redefine the threshold range and reconstruct.

[0094] Among them, in order to ensure the generation effect of the entire three-dimensional model, the "KeepLargest" option and the "Fill Holes" option need to be checked during the threshold cutting process.

[0095] "Keep Largest" is an operation of a functional nature used to screen and retain the largest object or structure during the processing of a three-dimensional model. This function automatically calculates the volume or size of each object or structure and selects the part with the largest volume. The remaining parts will be deleted or excluded from the processing scope. This is very useful for screening areas of interest or reducing the complexity of the model. "Fill Holes" is also an operation of a functional nature used to fill holes or vacancies in a three-dimensional model. It uses algorithms and interpolation techniques to infer missing data and fill in blank areas. The filling result will be smoothed and fused according to the surrounding geometric shape and data structure to ensure that the filled model is connected to the surrounding structure and has a consistent appearance. This function is very useful in medical image processing, engineering design, and other applications that require a complete model. By filling holes, the shape of the model can be analyzed more accurately, measurements and calculations can be performed, and subsequent model operations and analyses can be carried out.

[0096] The three-dimensional carotid artery model obtained by threshold segmentation in this embodiment is as Figure 4 shown.

[0097] S23. Based on step S22, perform a cut on the extracted three-dimensional blood vessel model to find the required carotid artery edge part and intercept it;

[0098] The cutting operation is still carried out in Mimics. Use the "Orthogonal to Screen" function in the 3D tool to create a cross-section perpendicular to the screen view on the three-dimensional model. At the same time, the segmented shape is presented in a different color, and the model before cutting is retained. During the operation, choose to continue cutting or recut according to the segmentation effect, ensuring the rigor and efficiency in graphic segmentation.

[0099] The comparison diagram of blood vessel cutting effect is as Figure 5 shown below.

[0100] In this embodiment, step S3 is specifically as follows:

[0101] S31. For the carotid artery edge intercepted in step S23, intercept it again, cut and remove the disordered entities similar to noise generated due to image reasons, and retain the model of the lesion part;

[0102] S32. Based on step S31, perform overall surface smoothing on the model of the lesion part and export it in stl format;

[0103] After model cutting, uneven faults will appear in some places, and there are also some places where the number of slices of the imported DICOM file is small, resulting in the absence of some cavities or faults. Although the option of filling cavities is selected when forming the mask, there are still some uneven situations in some positions, and overall smoothing processing needs to be performed in Mimics. Smoothing processing can reduce or remove the geometric noise, sharp edges or irregular shapes on the model surface, so that the appearance of the model is smoother and more continuous.

[0104] For the cavities and faults missing caused by the small number of file slices and the geometric noise, sharp edges or irregular shapes on the model surface, perform overall smoothing processing on the entire separately intercepted lesion part.

[0105] By adjusting the comparison iteration times and the smooth factor parameters, adjust the effect of overall smoothing processing. In this embodiment, under the condition that the smooth factor "Smooth Factor" is "0.4", perform smooth iteration successively to obtain the effect after multiple smooth processing iterations, as Figure 6 shown below.

[0106] The overall smoothing processing should not affect the entire model parameters, and some overly prominent cavities, fault missing sharp edges or irregular shapes are processed by step S33.

[0107] S33. Import the carotid artery lesion model that has been surface-smooth processed in step S32 into 3-matic (software for three-dimensional analysis model) and perform local defect smoothing processing;

[0108] "Local processing" is a function used to perform local modification, editing or analysis on a specific area of a three-dimensional model. Perform various operations on the selected part without affecting the entire model. As Figure 7As shown, for the defects such as sharp edges or irregular shapes that still exist after overall smoothing, local smoothing operations are performed on the defective parts, and local smoothing and trimming are carried out without affecting other model areas.

[0109] S34. Detect whether there are defects in the model processed in step S33. If there are defects, continue to repair them. If there are no defects, enter step S4;

[0110] Among them, the parameters for detection include: Full analysis, Inverted normals, Bad edges, Near bad edges, Planar holes, Shells, Overlapping triangles, Intersecting triangles.

[0111] In this embodiment, step S4 is specifically as follows:

[0112] S41. Perform end face resection on the repaired model, and name the cutting surface and the carotid artery surface;

[0113] Based on step S3, after completing the overall surface and local smoothing, the original cross-section has been deformed. Then, cut again in the direction perpendicular to the blood vessel. The newly generated cross-section is used to define the surface for input and output boundary conditions, and name it for convenient selection and condition definition during subsequent hemodynamic simulation.

[0114] Among them, for the naming of the boundaries, as shown in Table 1, the cross-section of the common carotid artery where blood flows in is set as in, and the cross-sections of the internal carotid artery and external carotid artery where blood flows out are set as out1 and out2 respectively, and the blood vessel wall is set as wall.

[0115] Table 1

[0116] Name Represented position in Common carotid artery (input) out2 External carotid artery (output) out1 Internal carotid artery (output) wall Vessel wall

[0117] S42. Based on step S41, divide the surface mesh and volume mesh of the model under appropriate parameters;

[0118] The meshes used are surface meshes and volume meshes. They are used to convert the geometric body into a discrete mesh representation form. "Surface Meshing" and "Volume Meshing" are two different meshing methods, which are respectively suitable for processing the meshing of geometric surfaces and the interior of geometric bodies.

[0119] For surface meshes, perform triangular mesh generation or modification operations in "Target triangle edge length", and the item to be modified is the desired size of the triangle edge length. At the same time, set parameters, including: "Preserve surface contours", "Maximum triangle edge length", "Influence area", "Growth rate".

[0120] Similarly, the parameters set in the volume mesh are the same as those in the surface mesh. The settings of the surface mesh and the volume mesh parameters in this embodiment are as Figure 8 Euclidean.

[0121] Among them, in the setting of the mesh parameters in this embodiment, select the element type as tetrahedron, set the maximum edge length to 1.0000, the growth rate to 25.0000, the aspect ratio of the shape measure to 100.0000, the influence area to the entire model, and add and modify the remaining parameters according to the actual situation.

[0122] S43. As Figure 9 shown, perform defect detection on the model after meshing in step S42 again, and modify it until there are no defects;

[0123] S44. Export the model with defect-free divided surface and volume meshes, and the format is.mesh.

[0124] In this embodiment, the specific steps of step S5 are as follows:

[0125] S51. Import the mesh file obtained in step S4 into ANSYS software (computational fluid dynamics software) and check the relevant parameters of the model size, surface, and volume meshes, and select the simulation model related to fluid mechanics;

[0126] In this embodiment, select "Viscous" and then "k-epsilon model" in "Models". The k-epsilon model is a commonly used turbulence model for simulating the turbulent behavior of viscous fluids. Set the parameters of the blood, including: blood density: 1060 kg / m³, blood viscosity: 0.003 (this value is for normal blood parameters, and corresponding adjustments can be made for special blood parameters).

[0127] S52. Define the boundary conditions for the end face and surface named in step S4;

[0128] In the three-dimensional model reconstruction stage, only four boundaries appear in the model with good cutting and smoothing effects, namely the four surface positions named in step S41. Select the input type of the common carotid artery as velocity input, select the output types of the internal carotid artery and the external carotid artery as pressure output, and input the blood flow velocity parameters detected by the actual device. The setting effect is as Figure 10 shown.

[0129] S53. Add cross-sections with the stenosis site and its front and back as key positions to detect data;

[0130] Detect the velocity for the entire inflow and outflow and compare various parameters at the stenosis before and after the operation. Then create planes parallel to the plane where the xy-axis is located, with positions in front of the carotid artery stenosis position, at the carotid artery stenosis position, and behind the carotid artery stenosis position, and name them "before", "narrow", and "after" in sequence to reflect the differences in the same parameters at different positions in a comparative manner. The positions in the model are as Figure 11 shown.

[0131] S54. Select an appropriate residual value according to the accuracy requirements of the pre-set experimental data for simulating the simulation results;

[0132] S55. Based on step S54, export the simulation results in the formats of.cas and.dat files.

[0133] In this embodiment, step S6 is specifically as follows:

[0134] S61. Check the imported.cas and.dat files and screen various parameter attempts;

[0135] S62. Select the exported part and exported parameters;

[0136] S63. Export the parameters of the two simulation results before and after the operation in the three-dimensional picture comparison format;

[0137] The selected parameters include: streamline diagram of flow velocity, wall shear stress, and blood vessel pressure. Adjust the data to be exported according to actual needs. In this embodiment, the default is to fully characterize and output single-item data to obtain the best result feedback. If multiple data superposition representation is required, adjust the parameter ratio according to the actual situation.

[0138] For the streamline diagram of flow velocity: Right-click on "streamline" in the upper operation interface, name it "streamline", select "in" in the "Start From" interface, fill in "100" at "of Points", and export the visualization result of this item of data. The comparison diagram of the velocity streamline parameters after preoperative and postoperative visualization processing exported in this embodiment is asFigure 12 as shown

[0139] Cloud map of Wss data: Right-click on "Contour" cloud map in the upper operation interface, name it "wss", then select all positions in the "Locations" interface, select "wall shear" in "Variable". Enter "100" in "ofcontours". Uncheck "Lighting" in "render", and export the visualization result of this item of data. The comparison chart of wall shear stress parameters after preoperative and postoperative visualization processing exported in this embodiment is as Figure 13 shown

[0140] Cloud map of Pressure data: Right-click on "Contour" cloud map in the upper operation interface, name it "pressure", then select all positions in the "Locations" interface, select "pressure" in "Variable". Enter "100" in "ofcontours", uncheck "Lighting" in "render", and export the visualization result of this item of data. The comparison chart of vascular pressure parameters after preoperative and postoperative visualization processing exported in this embodiment is as Figure 14 shown

[0141] S64. Export the detailed conditions of each position as a scatter plot and export the corresponding result data set;

[0142] In this embodiment, for the exported data, check "Write to File", select the installation address, and the data is saved as a.txt file in numerical form. You can also select the screenshot button on the side of the interface to save it in the form of a picture. Figure 15 are the velocity parameters before the stenosis position, at the stenosis position, and after the stenosis position.

[0143] For the pressure parameters in each position direction applied to other aspect calculations, select to save them in.txt format.

[0144] The parameters exported with the X-axis as the coordinate axis include: "Static Pressure", "Pressure Coefficient", "Dynamic Pressure", "Absolute Pressure", "Total Pressure". The parameters exported with the Y-axis as the coordinate axis include: "Static Pressure", "Wall Whear Stress", "Turbulent Kinetic Energy".

[0145] In summary, the method of the present invention utilizes clinical CTA data, performs three-dimensional reconstruction through the image processing software MIMICS, constructs a model including the common carotid artery, the carotid stenosis, the internal carotid artery, and the external carotid artery, and conducts finite element mesh division on the model, and finally performs hemodynamic analysis in ANSYS. This personalized model can more realistically reflect the physiological characteristics of patients, provide an effective tool for analyzing various parameters and changes at the carotid stenosis before and after surgery, assist doctors in making more scientific clinical decisions by providing visual and detailed analysis results, improve the treatment effect of patients, provide detailed guidance for novices, make it easier for them to get started with relevant software and technologies, and promote the development of academic and clinical applications.

[0146] The above are only the embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. An image processing and simulation analysis method for carotid artery images, the specific steps are as follows: S1. CTA image acquisition and processing: import CTA images into medical image reading software for observation and select appropriate CTA image files; First, obtain CTA images, that is, the contrast agent is injected intravenously and carried with the blood to the blood vessel area that needs to be imaged. X-rays pass through the patient's body and interact with the contrast agent in the blood vessels to form images of the blood vessels. Then, find the images with high contrast and many slices after adding contrast agents as files, and record their file names and numbers; in, The acquired CTA images, namely CTA image files, are discontinuous layered files in .dcm format generated by CTA technology, including: brain CT perfusion imaging and brain CT vascular imaging of patients with carotid artery stenosis; Then import the CTA image files for observation one by one, find the images with high vascular contrast and many slice layers after adding contrast agent as files, record their file names and numbers, and use this type of CTA image files as original files; S2, 3D model reconstruction, import the original CTA image selected in step S1, adjust the image contrast, select the case with the largest contrast, extract the blood vessel part, generate a 3D model and perform a cut to intercept the edge of the carotid artery; S3, model optimization, extracting carotid artery lesions, and performing overall and local surface optimization processing on the carotid artery lesion model; S4, mesh generation, performing local surface cutting, naming and meshing of the model processed in step S3, and then checking and correcting the surface; S5, CFD hemodynamic simulation, importing the grid file, setting the corresponding parameters, and performing simulation; S6. Based on the simulation results obtained in step S5, data analysis and visualization are performed, that is, the simulation results are visualized and the pre-operative and post-operative data are compared and analyzed.

2. The image processing and simulation analysis method for carotid artery images according to claim 1, characterized in that: The step S2 is specifically as follows: S21, import the original CTA image file and adjust the image contrast; Import the original file selected in step S1 into the medical image processing software, adjust the grayscale specification to "Soft tissue scale", enhance the contrast of the medical image in the software, and obtain the required carotid artery; S22, based on step S21, finding the area where the carotid artery is located, marking it by threshold comparison, and extracting a three-dimensional model of the blood vessel; After adjusting the contrast of the image, the blood vessel part represented by the highlighted area is extracted from the two-dimensional image by threshold segmentation and converted into a three-dimensional model; the lower limit of the threshold is adjusted in the image, and the selected content is the brightness between the lower limit of the threshold and the upper limit of the interval; after the threshold range is determined, a mask with operation is formed; the mask is calculated to construct the overall three-dimensional model, and the three sections are compared to see whether there is a difference with the three-dimensional model. If there is a difference, the threshold range is redefined and reconstructed; Among them, in the process of threshold cutting, check the "Keep Largest" option and the "Fill Holes" option; through "Keep Largest", filter and retain the largest objects or structures; through "Fill Holes", fill the holes or gaps in the 3D model and smooth and merge them according to the surrounding geometric shapes and data structures; S23, based on step S22, cutting the extracted three-dimensional blood vessel model once, finding the required edge of the carotid artery, and intercepting it; The cutting operation is still performed in the medical image processing software. Use the "Orthogonal to Screen" function in the 3D tool to create a section perpendicular to the screen view on the 3D model. At the same time, the segmented shape is displayed in other colors, and the model before cutting is retained. During the operation, choose to continue cutting or re-cut according to the effect after segmentation.

3. The image processing and simulation analysis method for carotid artery images according to claim 1, characterized in that: The step S3 is specifically as follows: S31, further intercepting the edge of the carotid artery intercepted in step S23, cutting and removing disordered entities similar to noise generated due to image reasons, and retaining the model of the lesion part; S32, based on step S31, the overall surface of the model of the lesion part is smoothed and exported into the stl format; For the holes and missing faults caused by the small number of file slices, as well as the geometric noise, sharp edges or irregular shapes on the model surface, the entire separately intercepted lesion part is smoothed as a whole; By adjusting the comparison iteration times and the smoothing factor parameters, the overall smoothing effect is adjusted. Under the condition of appropriate smoothing factors, smoothing iterations are performed one by one to obtain the effect after multiple smoothing iterations. The suitable condition of the smoothing factor is that each time under the smoothing factor parameter, each smoothing operation iteration will cause an image but will not significantly change the model; S33, importing the carotid artery lesion model after the surface smoothing process in step S32 into the software of the three-dimensional analytical model, and performing smoothing process on the local defects; For defects such as sharp edges or irregular shapes that still exist after overall smoothing, local smoothing operations are performed on the defective parts without affecting other model areas. S34, check whether the model processed in step S33 has defects, if yes, continue to repair, if no, proceed to step S4; The detection parameters include: Full analysis, Inverted normals, Bad edges, Near badedges, Planar holes, Shells, Overlapping triangles, and Intersecting triangles.

4. The image processing and simulation analysis method for carotid artery images according to claim 1, characterized in that: The step S4 is specifically as follows: S41. Perform end-surface resection on the repaired model and name the cut surface and the carotid artery surface; Based on step S3, after the entire surface and local smoothing are completed, if the original cross section has been deformed, it is cut again in the direction perpendicular to the blood vessel. The newly generated cross section is used to define the input and output boundary conditions, and is named; Among them, for the naming of boundaries, the section of the common carotid artery that flows into the blood is set as in, the sections of the internal carotid artery and external carotid artery that flow out of the blood are set as out1 and out2 respectively, and the blood vessel wall is set as wall; S42, based on step S41, dividing the model into surface mesh and volume mesh under appropriate parameters; For surface meshes, generate or modify the triangle mesh in "Target triangle edge length", and the modified item is the desired triangle edge length; at the same time, set parameters, including: "Preserve surface contours", "Maximum triangle edge length", "Influence area", "Growth rate"; Similarly, the parameters set in the volume mesh are consistent with those in the surface mesh; S43, performing defect detection again on the model after meshing in step S42, and modifying it until no defects exist; S44. Export the defect-free divided surface and volume mesh model in .mesh format.

5. The image processing and simulation analysis method for carotid artery images according to claim 1, characterized in that: The step S5 is specifically as follows: S51, importing the grid file obtained in step S4 into computational fluid dynamics software and checking the relevant parameters of the model size, surface and volume grids, and selecting a simulation model related to fluid mechanics; The k-epsilon model is used as the simulation model to set the blood parameters, including: blood density, blood viscosity; S52, defining boundary conditions for the end face and the surface named in step S4; In the three-dimensional model reconstruction stage, the model with good cutting and smoothing effects only has four boundaries, namely, the four surface positions named in step S41; the input type of the common carotid artery is selected as velocity input, the output type of the internal carotid artery and the external carotid artery is selected as pressure output, and the blood flow velocity parameters detected by the actual device are input; S53, adding cross sections around the stenosis site and its front and back as key locations to detect data; The speed of the entire inflow and outflow is detected, and the parameters at the stenosis are compared before and after the operation. A plane parallel to the plane where the xy axis is located is created, and the positions are respectively before the carotid stenosis position, the carotid stenosis position, and after the carotid stenosis position, and they are named "before", "narrow", and "after" in turn to reflect the difference of the same parameter at different positions by comparison. S54, according to the accuracy requirements of the pre-set experimental data, selecting a suitable residual value to simulate the simulation results; S55. Based on step S54, the simulation results are exported in the format of .cas and .dat files.

6. The image processing and simulation analysis method for carotid artery images according to claim 1, characterized in that: The step S6 is specifically as follows: S61, checking the imported .cas and .dat files, and screening various parameter views; S62, selecting the exported part and export parameters; S63, exporting parameters of the two simulation results before and after the operation in a three-dimensional picture comparison format; The selected parameters include: flow velocity streamline diagram, surface shear force and vascular pressure; the data to be exported are adjusted according to actual needs, and the single data is fully characterized and output by default to obtain the best result feedback. If multiple data need to be superimposed, the parameter weight is adjusted according to the actual situation; S64, exporting the details of each position in a scatter plot, and exporting a corresponding result data set; The data is saved as a .txt file in numerical form, and the screenshot button on the side of the interface is selected to save it in the form of a picture, and it can be adjusted according to actual needs; For the pressure parameters in various positions and directions used for other calculations, choose to save them in .txt format; The parameters derived from the X-axis include: "Static Pressure", "Pressure Coefficient", "Dynamic Pressure", "Absolute Pressure", and "Total Pressure"; the parameters derived from the Y-axis include: "Static Pressure", "Wall Whear Stress", and "Turbulent Kinetic Energy".

Images