Augmented reality vascular fusion navigation system for transcatheter occlusion

Abstract

The present application relates to the technical field of medical navigation, in particular to an augmented reality blood vessel fusion navigation system for transcatheter occlusion, comprising a real-time image acquisition module, a blood vessel three-dimensional model construction module, an image enhancement module, a navigation module and a fusion module, the image acquisition module is connected with an ultrasound catheter to acquire blood vessel three-dimensional images, the model construction module creates a blood vessel three-dimensional model based on the images, the image enhancement module enhances the blood vessel images through spatial adaptive transformation, key point recognition, region adaptive segmentation and edge feature extraction, the navigation module determines the catheter tip position by using the blood vessel three-dimensional model and the enhanced images, the fusion module fuses the enhanced images with the blood vessel model to generate fusion display data, and an AR device overlays and displays the blood vessel three-dimensional model at the catheter tip position based on the catheter tip position and the fusion data, which significantly improves the visualization effect of complex blood vessel structures and provides an accurate spatial reference for catheter navigation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical navigation, in particular to an augmented reality blood vessel fusion navigation system for transcatheter occlusion, which is applied to the interventional treatment of cardiovascular diseases such as atrial septal defect and ventricular septal defect. BACKGROUND

[0002] Transcatheter occlusion is an important minimally invasive surgical method for treating congenital heart diseases such as atrial septal defect and ventricular septal defect. Traditional transcatheter occlusion mainly relies on ultrasound and X-ray images for navigation, but these methods have problems such as poor image quality, lack of depth information, and excessive radiation exposure.

[0003] Most existing navigation systems use two-dimensional image display, which cannot accurately represent the three-dimensional structure of blood vessels. At the same time, the fusion degree of image data and blood vessel models is not high, which limits the accuracy of surgical navigation. In addition, the traditional system lacks effective image enhancement processing of blood vessel images, which makes it difficult to identify complex blood vessel structures and affects the accurate positioning and release of the occluder.

[0004] With the development of augmented reality (AR) technology, it is possible to apply AR technology to transcatheter occlusion navigation, but there is currently a lack of high-precision AR blood vessel fusion navigation system specifically for transcatheter occlusion, especially a lack of effective image enhancement technology and accurate spatial registration method, which limits the application effect of AR technology in this field. SUMMARY

[0005] The purpose of the present application is to provide an augmented reality blood vessel fusion navigation system for transcatheter occlusion, which realizes accurate enhancement of blood vessel images, accurate construction of three-dimensional models, and accurate navigation of catheters through the cooperation of multiple modules, thereby improving the safety and success rate of transcatheter occlusion.

[0006] The present application provides an augmented reality blood vessel fusion navigation system for transcatheter occlusion, which comprises:

[0007] A real-time image acquisition module connected to an ultrasound catheter for acquiring three-dimensional spatial images of blood vessels;

[0008] A blood vessel three-dimensional model construction module connected to the real-time image acquisition module for constructing a blood vessel three-dimensional model based on the three-dimensional spatial images of blood vessels;

[0009] An image enhancement module connected to the real-time image acquisition module for enhancing the three-dimensional spatial images of blood vessels, the image enhancement module comprising:

[0010] A spatial adaptive transformation unit for performing spatial adaptive transformation processing on the three-dimensional spatial images of blood vessels to generate a grayscale image;

[0011] A key point recognition and target tracking unit, connected with the spatial adaptive transformation unit, is configured to recognize blood vessel key points on the gray-scale image and construct a blood vessel connectivity structure;

[0012] A region adaptive segmentation unit, connected with the key point recognition and target tracking unit, is configured to calculate a blood vessel region information entropy and perform adaptive segmentation based on the blood vessel connectivity structure;

[0013] An edge feature extraction and splicing unit, connected with the region adaptive segmentation unit, is configured to extract edges of the segmented blood vessel region and generate a region feature image;

[0014] A multi-feature fusion optimization unit, connected with the edge feature extraction and splicing unit, is configured to perform feature selection, weighted fusion and optimization on the region feature image;

[0015] A navigation module, connected with the blood vessel three-dimensional model construction module and the image enhancement module respectively, is configured to perform fusion comparison based on the blood vessel three-dimensional model and the enhanced blood vessel three-dimensional space image, and determine a catheter tip position;

[0016] A fusion module, connected with the image enhancement module and the blood vessel three-dimensional model construction module, is configured to fuse the enhanced blood vessel three-dimensional space image and the blood vessel three-dimensional model to generate fusion display data;

[0017] An AR device, connected with the navigation module and the fusion module, is configured to superimpose display the blood vessel three-dimensional model at the catheter tip position based on the catheter tip position and the fusion display data.

[0018] Preferably, the spatial adaptive transformation unit comprises:

[0019] An image preprocessing component is configured to perform noise reduction and equalization processing on the blood vessel three-dimensional space image;

[0020] A local characteristic analysis component, connected with the image preprocessing component, is configured to calculate local region statistical characteristics of the blood vessel three-dimensional space image;

[0021] A transformation function construction component, connected with the local characteristic analysis component, is configured to construct a spatial adaptive transformation function based on the local region statistical characteristics;

[0022] A transformation execution component, connected with the transformation function construction component, is configured to apply the spatial adaptive transformation function to process the blood vessel three-dimensional space image;

[0023] A gray-scale conversion component, connected with the transformation execution component, is configured to convert the transformed blood vessel three-dimensional space image into a gray-scale image.

[0024] As preferred, the key point recognition and target tracking unit comprises:

[0025] a feature point detection component for detecting blood vessel feature points on the gray image;

[0026] a multi-scale analysis component connected with the feature point detection component for analyzing the blood vessel feature points in different scale spaces;

[0027] a connected structure construction component connected with the multi-scale analysis component for establishing a blood vessel network connected graph based on the blood vessel feature points;

[0028] a key point labeling component connected with the connected structure construction component for classifying and labeling the blood vessel feature points;

[0029] a dynamic target tracking component connected with the key point labeling component for establishing a tracking model to realize dynamic tracking of the blood vessel feature points.

[0030] As preferred, the region adaptive segmentation unit comprises:

[0031] an information entropy calculation component for calculating information entropy of each blood vessel unit in the blood vessel connected structure;

[0032] a sub-region division component connected with the information entropy calculation component for dividing each blood vessel unit into multiple sub-regions;

[0033] a region characteristic analysis component connected with the sub-region division component for analyzing complexity of different regions based on information entropy;

[0034] an adaptive threshold generation component connected with the region characteristic analysis component for automatically calculating a segmentation threshold according to region information entropy;

[0035] a region labeling component connected with the adaptive threshold generation component for labeling segmented regions.

[0036] As preferred, the edge feature extraction and splicing unit comprises:

[0037] an edge detection component for detecting blood vessel edges using gradient and direction information;

[0038] an edge enhancement component connected with the edge detection component for performing enhancement processing on the detected blood vessel edges;

[0039] a region component extraction component connected with the edge enhancement component for segmenting an image into multiple independent region components based on blood vessel edge information;

[0040] The feature image generation component is connected with the region component extraction component and is configured to generate a feature image for each region component.

[0041] The edge-guided stitching component is connected with the feature image generation component and is configured to guide accurate stitching of the region components by using edge information.

[0042] Preferably, the multi-feature fusion optimization unit comprises:

[0043] The feature analysis component is configured to analyze the feature image of each region component.

[0044] The feature screening component is connected with the feature analysis component and is configured to screen representative features based on information value evaluation.

[0045] The weight calculation component is connected with the feature screening component and is configured to calculate fusion weights according to feature importance and reliability.

[0046] The weighted fusion component is connected with the weight calculation component and is configured to integrate multiple features to generate an enhanced image by using a weighted fusion strategy.

[0047] The quality evaluation component is connected with the weighted fusion component and is configured to evaluate the quality of the fusion result and feed back the evaluation result to the aforementioned processing links.

[0048] Preferably, the blood vessel three-dimensional model construction module comprises:

[0049] The three-dimensional data acquisition unit is configured to read three-dimensional image data and perform three-dimensional spatial stereo reconstruction.

[0050] The blood vessel structure identification unit is connected with the three-dimensional data acquisition unit and is configured to acquire blood vessel trunks and blood vessel branches as well as starting points and ending points of the branches.

[0051] The blood vessel layering and dividing unit is connected with the blood vessel structure identification unit and is configured to divide the blood vessel trunks and the blood vessel branches into blood vessel segments and blood vessel layers.

[0052] The blood vessel connection processing unit is connected with the blood vessel layering and dividing unit and is configured to establish blood vessel segments connected between the blood vessel branches and the blood vessel trunks according to the starting points and the ending points of the blood vessels.

[0053] The blood vessel tree construction unit is connected with the blood vessel connection processing unit and is configured to merge the connected blood vessel segments to form a complete blood vessel tree structure.

[0054] Preferably, the navigation module comprises:

[0055] The registration processing unit is configured to calculate a spatial transformation relationship between the blood vessel three-dimensional model and the blood vessel three-dimensional spatial image after the enhancement processing.

[0056] a position tracking unit, connected to the registration processing unit, for tracking real-time position of the catheter tip in the three-dimensional space of the blood vessel;

[0057] a trajectory prediction unit, connected to the position tracking unit, for predicting future motion trajectory of the catheter tip based on historical position data of the catheter tip;

[0058] a space mapping unit, connected to the trajectory prediction unit and the registration processing unit, for mapping the position of the catheter tip to the corresponding position of the three-dimensional model of the blood vessel;

[0059] a navigation information generation unit, connected to the space mapping unit, for generating catheter operation suggestions and path planning information.

[0060] Preferably, the fusion module comprises:

[0061] a data synchronization unit for ensuring synchronization of the three-dimensional space image of the blood vessel after enhanced processing and the three-dimensional model of the blood vessel in time and space;

[0062] a feature matching unit, connected to the data synchronization unit, for identifying corresponding feature points in the two kinds of data;

[0063] a transparency adjustment unit, connected to the feature matching unit, for dynamically adjusting the transparency of the model according to clinical needs and image quality;

[0064] a view angle optimization unit, connected to the transparency adjustment unit, for optimizing the display view angle of the model according to the operator's line of sight and the need for surgery;

[0065] a fusion rendering unit, connected to the view angle optimization unit, for rendering the adjusted data into a final display image.

[0066] Preferably, the AR device comprises:

[0067] a display component for displaying the fused three-dimensional model of the blood vessel in three-dimensional form;

[0068] a camera component for capturing real-time scene images of the surgery;

[0069] a processing component, connected to the display component and the camera component, for integrating the scene images of the surgery and the three-dimensional model of the blood vessel;

[0070] an interaction component, connected to the processing component, for receiving the gestures or voice instructions of the doctor and adjusting the display content;

[0071] a communication component, connected to the processing component, for data exchange with the navigation module and the fusion module;

[0072] A positioning component connected with the processing component is configured to acquire the position and posture of the AR device in space in real time.

[0073] The present application has the following beneficial effects:

[0074] 1. Through the innovative image enhancement module, the adaptive enhancement of the blood vessel image is realized, and the visualization effect of the complex blood vessel structure is significantly improved;

[0075] 2. Through the accurate blood vessel three-dimensional model construction, the three-dimensional reconstruction of the main stem and branch of the blood vessel is realized, and accurate spatial reference is provided for catheter navigation;

[0076] 3. Through the combination of AR technology and accurate navigation, intuitive real-time positioning and path planning of the catheter are realized, and the surgical risk is reduced;

[0077] 4. The close cooperation between the modules of the system forms a closed-loop feedback mechanism, which improves the navigation accuracy and system reliability;

[0078] 5. The dependence on contrast agents and X-ray radiation exposure is reduced, and the surgical safety and the comfort of the doctor's operation are improved. BRIEF DESCRIPTION OF DRAWINGS

[0079] Figure 1 is the overall structure block diagram of the augmented reality blood vessel fusion navigation system for transcatheter occlusion surgery of the present application;

[0080] Figure 2 is the functional structure diagram of the image enhancement module of the present application;

[0081] Figure 3 is the working flowchart of the space adaptive transformation unit of the present application;

[0082] Figure 4 is the processing schematic diagram of the key point recognition and target tracking unit of the present application;

[0083] Figure 5 is the processing principle diagram of the region adaptive segmentation unit of the present application;

[0084] Figure 6 is the processing effect schematic diagram of the edge feature extraction and splicing unit of the present application;

[0085] Figure 7 is the processing flowchart of the multi-feature fusion optimization unit of the present application;

[0086] Figure 8 is the working flowchart of the blood vessel three-dimensional model construction module of the present application;

[0087] Figure 9 is the functional structure diagram of the navigation module of the present application;

[0088] Figure 10 is a processing schematic diagram of the fusion module of the present application. DETAILED DESCRIPTION

[0089] Reference is made to the accompanying Figures 1-10 The specific embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0090] Referring to Figure 1 The augmented reality blood vessel fusion navigation system for transcatheter occlusion provided by the present application comprises a real-time image acquisition module 1, a blood vessel three-dimensional model construction module 2, an image enhancement module 3, a navigation module 4, a fusion module 5 and an AR device 6.

[0091] The real-time image acquisition module 1 is connected with an ultrasonic catheter, and is used for acquiring a blood vessel three-dimensional space image. Specifically, the real-time image acquisition module 1 acquires a real-time image of a blood vessel structure inside a heart through a high-frequency ultrasonic probe, and the acquisition frequency is preferably 25-50 frames per second, and the resolution is preferably 512x512 pixels, so as to ensure that the time resolution and the spatial resolution of the image meet the clinical requirements.

[0092] The blood vessel three-dimensional model construction module 2 is connected with the real-time image acquisition module 1, and is used for constructing a blood vessel three-dimensional model based on the blood vessel three-dimensional space image. The blood vessel three-dimensional model construction module 2 constructs an accurate blood vessel three-dimensional model by processing a blood vessel image sequence acquired in real time, in combination with pre-acquired CT or MRI image data, so as to provide a spatial reference for navigation.

[0093] The image enhancement module 3 is connected with the real-time image acquisition module 1, and is used for performing enhancement processing on the blood vessel three-dimensional space image. The image enhancement module 3 is a key innovative point of the present system, and improves the blood vessel image quality through multi-stage processing, and enhances the visualization effect of the blood vessel structure.

[0094] The navigation module 4 is connected with the blood vessel three-dimensional model construction module 2 and the image enhancement module 3 respectively, and is used for performing fusion comparison based on the blood vessel three-dimensional model and the enhanced blood vessel three-dimensional space image, so as to determine the position of a catheter tip. The navigation module 4 realizes accurate positioning of the catheter in the blood vessel through an advanced registration algorithm and a position tracking technology.

[0095] The fusion module 5 is connected with the image enhancement module 3 and the blood vessel three-dimensional model construction module 2, and is used for fusing the enhanced blood vessel three-dimensional space image and the blood vessel three-dimensional model to generate fusion display data. The fusion module 5 realizes seamless integration of multi-source data, and generates unified visualization data.

[0096] The AR device 6 is connected with the navigation module 4 and the fusion module 5, and is used to superimpose and display the three-dimensional model of the blood vessel at the catheter tip position based on the catheter tip position and the fusion display data. The AR device 6 superimposes the virtual blood vessel model on the actual anatomic position of the patient through a perspective display technology, thereby providing the doctor with a "perspective" view.

[0097] Referring to Figure 2 , the image enhancement module 3 includes a spatial adaptive transformation unit 31, a key point recognition and target tracking unit 32, a region adaptive segmentation unit 33, an edge feature extraction and splicing unit 34, and a multi-feature fusion optimization unit 35, which are sequentially connected to form a processing pipeline.

[0098] Referring to Figure 3 , the spatial adaptive transformation unit 31 includes an image preprocessing component 311, a local characteristic analysis component 312, a transformation function construction component 313, a transformation execution component 314, and a gray scale conversion component 315.

[0099] The image preprocessing component 311 performs noise reduction and equalization processing on the three-dimensional spatial image of the blood vessel. Specifically, a bilateral filtering algorithm is used for noise reduction, which retains edge information while effectively suppressing noise. For an ultrasound image, an adaptive histogram equalization method is used for contrast enhancement, and the algorithm is described as follows:

[0100] First, the input image I is divided into several blocks of equal size, and the block size is preferably 8x8 or 16x16 pixels. For each block, its histogram is calculated and histogram equalization is applied to obtain the conversion function:

[0101] ,

[0102] wherein: is the conversion function of the kth image block, i is the pixel gray value (ranging from 0 to 255), is the number of pixels with a gray value of j, and N is the total number of pixels in the block.

[0103] Then, the conversion functions of the blocks are combined using a bilinear interpolation method to avoid discontinuity at the block boundaries:

[0104] ,

[0105] wherein: is the gray value of the output image at position (x, y), is the gray value of the input image at position (x, y), is the weight coefficient of the kth adjacent block, and satisfies .

[0106] The local property analysis component 312 calculates the local region statistical properties of the blood vessel three-dimensional space image. Specifically, the local statistical properties of each region of the image are calculated using a sliding window method, including local mean, variance, gradient amplitude and direction, etc. For the blood vessel image, the local variance can effectively reflect the contrast difference between the blood vessel and the background, so the local variance is calculated as the focus:

[0107] ,

[0108] wherein: is the local variance at position (x, y), represents the texture complexity of the image at the position, and W is the window size (usually 7x7 or 9x9 pixels when processing the cardiac ultrasound image), is the pixel gray value at position (i, j), is the local mean at position (x, y).

[0109] The transformation function construction component 313 constructs a spatial adaptive transformation function based on the local region statistical properties. The present application adopts an innovative adaptive transformation function, which automatically adjusts the transformation parameters according to the local properties:

[0110] ,

[0111] wherein: is the transformation function, which maps the original image gray value to the enhanced gray value, is the gray value of the original image at position , and are position-dependent adaptive parameters.

[0112] The two parameters are dynamically adjusted according to the local variance:

[0113] ,

[0114] ,

[0115] wherein: and are basic parameters, and the empirical values in the processing of the cardiac ultrasound image are 1.0 and 3.0, respectively, and are adjustment coefficients, and the empirical values are 0.5 and 0.8, respectively, is the maximum local variance in the image.

[0116] The transformation execution component 314 applies the spatial adaptive transformation function to process the blood vessel three-dimensional space image. Specifically, the transformation function is applied to each pixel in the image to achieve spatial adaptive enhancement processing:

[0117] ,

[0118] wherein: is the transformed image.

[0119] The gray scale conversion component 315 converts the transformed blood vessel three-dimensional space image into a gray scale image. For a color image, the conversion into a gray scale image is realized by weighted average:

[0120] ,

[0121] wherein: is the converted gray scale image, , and are the red, green and blue channel values of the original image at the position .

[0122] The innovation of the spatial adaptive transformation unit 31 lies in that, through local characteristic analysis and adaptive parameter adjustment, spatial adaptive enhancement of the blood vessel image is realized, and the visibility of the blood vessel structure is effectively improved, especially the enhancement effect for low contrast regions is remarkable.

[0123] Referring to Figure 4 , the key point recognition and target tracking unit 32 includes a feature point detection component 321, a multi-scale analysis component 322, a connected structure construction component 323, a key point marking component 324 and a dynamic target tracking component 325.

[0124] The feature point detection component 321 detects blood vessel feature points on the gray scale image. The present application adopts an improved FAST (Features from Accelerated Segment Test) corner detection algorithm, which has the characteristics of high computational efficiency and strong noise resistance, and is particularly suitable for real-time blood vessel image processing. The algorithm steps are as follows:

[0125] For each point p in the image, 16 pixel points around it with a radius of r (in blood vessel image processing, r is preferably 3) are investigated. If there are n consecutive points (n is preferably 9) whose gray scale values are all higher than the gray scale value of p by a threshold t or are all lower than the gray scale value of p by a threshold t, then the point p is determined as a corner point. The threshold t is adaptively set according to the statistical characteristics of the image:

[0126] ,

[0127] wherein: is the average gray scale value of the image, representing the overall brightness level, is the standard deviation of the image, representing the overall contrast level, and k is an adjustment coefficient, whose empirical value is 0.5.

[0128] The multi-scale analysis component 322 analyzes vascular feature points at different scales. By constructing the scale space of the image and detecting feature points at different scales, it helps to capture vascular structures of varying thicknesses.

[0129] ,

[0130] in: For the scale Scale-space images, The standard deviation is The Gaussian kernel function, where * denotes convolution operation.

[0131] In transcatheter occlusion, 3-5 scales are typically used, with scale parameters $\sigma$ of 1.0, 2.0, 4.0, 8.0 and 16.0 pixels, covering different structures from small vessels to major vascular trunks.

[0132] Component 323, which constructs a connected structure, builds a vascular network connectivity graph based on vascular feature points. An improved minimum spanning tree algorithm is used, with feature points as nodes, to construct the vascular connectivity structure.

[0133] First, calculate the Euclidean distance between feature points as the edge weights:

[0134] ,

[0135] in: For feature points and The edge weights represent the spatial distance between two points. and These are the coordinates of two feature points.

[0136] Then, Kruskal's algorithm is applied to construct the minimum spanning tree:

[0137] 1. Sort all edges by weight from smallest to largest; 2. Add edges one by one to avoid forming cycles; 3. Repeat until all nodes are connected; finally, optimize the generated connected graph based on vascular anatomy knowledge and remove unreasonable connections.

[0138] Keypoint labeling component 324 classifies and labels blood vessel feature points. Based on their topological location in the connected graph, the feature points are classified as bifurcation points, intersection points, and endpoints:

[0139] Bifurcation point: A node with a connectivity of 3 (where three blood vessels intersect);

[0140] Intersection: A node with a connectivity greater than 3 (where multiple blood vessels intersect);

[0141] Endpoint: A node with a connectivity of 1 (the end of a blood vessel);

[0142] In order to improve the classification accuracy, the local gray distribution and gradient information are combined for verification, and the false feature points are eliminated.

[0143] The dynamic target tracking component 325 establishes a tracking model to realize dynamic tracking of the blood vessel feature points. The improved KLT (Kanade-Lucas-Tomasi) optical flow tracking algorithm is adopted to realize tracking of the feature points between continuous frames:

[0144] Suppose that the gray value of the point (x, y) remains unchanged between time t and t+Δt, then:

[0145] ,

[0146] The displacement vector (Δx, Δy) is solved by least square optimization in the local window:

[0147] ,

[0148] Wherein, W represents the local window centered on the feature point, and the window size is preferably 15x15 pixels.

[0149] In order to improve the tracking stability, the pyramid LK optical flow algorithm is adopted, the image pyramid is constructed, and the solution is performed layer by layer from low resolution to high resolution, so that the large displacement condition is effectively handled.

[0150] The innovation of the key point recognition and target tracking unit 32 lies in that the accurate representation of the blood vessel network is realized through multi-scale analysis and topological structure construction, and the continuous tracking of the key blood vessel structure during the catheter operation is ensured through dynamic target tracking, thereby providing a reliable foundation for the region segmentation.

[0151] Referring to Figure 5 , the region adaptive segmentation unit 33 includes an information entropy calculation component 331, a sub-region division component 332, a region characteristic analysis component 333, an adaptive threshold generation component 334, and a region marking component 335.

[0152] The information entropy calculation component 331 calculates the information entropy of each blood vessel unit in the blood vessel connected structure. The information entropy is an important index for measuring the complexity of the image region, and the gray information entropy calculation method is adopted in the present application:

[0153] ,

[0154] Wherein, H is the information entropy, L is the gray level (usually 256), and p(i) is the probability of the pixel with the gray value i appearing in the region.

[0155] The sub-region division component 332 divides each blood vessel unit into multiple sub-regions. Considering the continuity and variability of blood vessel structure, the blood vessel unit is divided into multiple sub-regions along the blood vessel direction, and the number of sub-regions is preferably 5-8 to balance the calculation complexity and accuracy requirement.

[0156] For each sub-region, its local information entropy is calculated:

[0157] ,

[0158] Wherein: is the information entropy of the i-th sub-region, is the probability of the pixel with the gray value of k appearing in the sub-region.

[0159] The region characteristic analysis component 333 analyzes the complexity of different regions based on the information entropy. The application innovatively introduces the concept of weighted information entropy, which comprehensively considers the information entropy and position importance of the sub-region:

[0160] ,

[0161] Wherein: represents the information entropy of the i-th blood vessel unit, represents the information entropy of the j-th sub-region of the i-th blood vessel unit, represents the number of sub-regions of the i-th blood vessel unit, represents the target weight value, which is determined according to the importance of the sub-region in the blood vessel structure, and the preferred range is 0.5-2.0. The local entropy value is calculated to reflect the entropy difference between regions: ,

[0162] Wherein:

[0163] represents the local entropy value, which represents the difference degree of the region from the average level, represents the average value of the information entropy of all blood vessel units,

[0164] The adaptive threshold generation component 334 automatically calculates the segmentation threshold according to the region information entropy. The traditional fixed threshold segmentation method is difficult to adapt to the change of complex blood vessel structure, and the application dynamically generates the segmentation threshold based on the information entropy: ,

[0165] Wherein:

[0166] is the i-th sub-region, is the j-th sub-region, is the k-th sub-region.

[0167] ​​​​Segmentation threshold of a blood vessel unit, is a basic threshold (usually set as 0.85 times of the average gray value of the image in the cardiac ultrasound image), is an adjustment coefficient, and the empirical value is 0.3.

[0168] The region marking component 335 marks the segmented regions. The connected component analysis method is used to mark and analyze the connectivity of the segmentation result, so as to eliminate isolated small regions and noise interference:

[0169] For each connected region if the area thereof is less than a preset threshold (the empirical value is 50 pixels), the connected region is merged into a neighboring large region or deleted. In this way, small regions caused by noise can be effectively removed, and the integrity of the blood vessel structure is maintained.

[0170] The innovation of the region adaptive segmentation unit 33 lies in that the accurate segmentation of the complex blood vessel structure is realized through the adaptive segmentation mechanism driven by information entropy, and especially for the regions with large contrast variation, the excellent adaptive ability is shown. This lays a foundation for the subsequent edge extraction and feature fusion.

[0171] Referring to Figure 6 , the edge feature extraction and splicing unit 34 includes an edge detection component 341, an edge enhancement component 342, a region component extraction component 343, a feature image generation component 344, and an edge guided splicing component 345.

[0172] The edge detection component 341 detects the blood vessel edge by using gradient and direction information. The improved Canny edge detection algorithm is used in the application, and the algorithm is optimized according to the characteristics of the blood vessel structure:

[0173] Firstly, a Gaussian filter is used to smooth the image:

[0174] ,

[0175] wherein: is the smoothed image, is a Gaussian kernel function, is a smoothing parameter, and the empirical value is 1.2.

[0176] Then, the image gradient is calculated:

[0177] ,

[0178] ,

[0179] wherein, and gradients in x and y directions, respectively, represent the rate of change of the image at the point

[0180] Gradient magnitude and direction:

[0181]

[0182]

[0183] Finally, edges are obtained by non-maximum suppression and double thresholding:

[0184] Non-maximum suppression is performed to keep the local maximum points in the gradient direction.

[0185] Double thresholding and edge linking are used, with low threshold and high threshold set to 15% and 30% of the maximum gradient magnitude, respectively.

[0186] The edge enhancement component 342 performs enhancement processing on the detected vessel edges. To improve the continuity and clarity of the edges, the present application adopts an adaptive edge enhancement method:

[0187]

[0188] wherein: is the enhanced edge image, is the original edge image, is the edge continuity measure, is the enhancement coefficient, with an empirical value of 0.6.

[0189] The edge continuity measure is calculated based on the local edge direction consistency:

[0190]

[0191] wherein: W is the window size, preferably 5x5 pixels.

[0192] The region component extraction component 343 segments the image into multiple independent region components based on the vessel edge information. Using the edges as the segmentation boundary, a region growing algorithm is used to extract each region component:

[0193] Starting from the seed point of each region, the region is gradually expanded until it reaches the edge or meets the termination condition. For a catheter occlusion vessel image, the region is usually divided into different components such as the vessel lumen, vessel wall, and surrounding tissue.

[0194] The feature image generation component 344 generates a feature image for each region component. For each region component, its feature image is extracted to preserve the unique features of the region:

[0195] ​​​​ ,

[0196] in: The feature image of the k-th region component. For the k-th region, such as a defective region, atrial septal tissue, etc.

[0197] Furthermore, the feature image is enhanced to highlight regional characteristics:

[0198] ,

[0199] in For the enhanced feature image, The average gray value of the region. The preferred range for the enhancement factor is 0.3-0.8.

[0200] Edge-guided stitching component 345 utilizes edge information to guide the precise stitching of regional components. Traditional image stitching methods may lead to distortion at the edges. This invention innovatively proposes an edge-guided stitching method:

[0201] ,

[0202] in: The image is stitched together, where K is the total number of region components. For position The weight of the component in the k-th region.

[0203] The weighting function is defined based on the distance to the edge:

[0204] ,

[0205] in: For position To the The distance between the boundaries of each region The smoothing parameter has an empirical value of 3.0.

[0206] The innovation of the edge feature extraction and stitching unit 34 lies in the fact that, through edge-guided feature extraction and stitching, it achieves accurate separation and seamless integration of features from different regions in the blood vessel image, especially preserving the fine structure at the edge, and providing a high-quality feature image for subsequent feature fusion.

[0207] See Figure 7 The multi-feature fusion optimization unit 35 includes a feature analysis component 351, a feature screening component 352, a weight calculation component 353, a weighted fusion component 354, and a quality evaluation component 355.

[0208] The feature analysis component 351 analyzes the feature images of each region component. For each feature image, multiple feature descriptors are extracted, including statistical features, texture features, and shape features:

[0209] Statistical features: mean, variance, skewness, kurtosis, etc.

[0210] Texture features: energy, contrast, correlation, etc. based on gray level co-occurrence matrix

[0211] Shape features: edge density, direction consistency, etc.

[0212] The feature selection component 352 selects representative features based on information value evaluation. The feature selection method based on information gain is adopted to remove redundant features and retain the most representative features:

[0213] For a feature , its information gain is calculated:

[0214] ,

[0215] Where: is the information gain of feature , is the information entropy of the region, is the conditional entropy of the region given the known feature .

[0216] The top N features with the highest information gain are selected, N is determined according to the computing resources and accuracy requirements. For real-time processing of transcatheter occlusion, N is preferably 5-8.

[0217] The weight calculation component 353 calculates the fusion weight according to the feature importance and reliability. The adaptive weight calculation method is adopted in the present application, which comprehensively considers the information gain, region stability and edge sharpness of the feature:

[0218] ,

[0219] Where: is the fusion weight of feature , is the stability measure of the region where feature is located, is the edge sharpness measure of the region where feature is located.

[0220] The stability measure is defined based on the variance of the pixel values in the region:

[0221] ,

[0222] Where: is the pixel variance of the region where the feature is located, is the baseline variance, and the experience value is 0.5 times the global variance of the image.

[0223] The edge sharpness metric is defined based on the edge gradient magnitude:

[0224]

[0225] wherein: is the set of boundary pixels of the region where the feature is located, is the boundary length, is the gradient magnitude at position

[0226] The weighted fusion component 354 integrates multiple features to generate the enhanced image using a weighted fusion strategy. The feature images are fused by weighting according to the calculated weights:

[0227]

[0228] wherein: is the fused image, is the i-th feature image.

[0229] To maintain the overall visual effect of the image, the fusion result is normalized and adjusted in grayscale:

[0230]

[0231] wherein: is the final enhanced image, and are the grayscale adjustment parameters, which are ensured to be in the range of [0, 255] by linear stretching.

[0232] The quality evaluation component 355 evaluates the quality of the fusion result and feeds back the evaluation result to the aforementioned processing links. The present application uses a multi-index comprehensive evaluation method to evaluate the quality of the enhanced image:

[0233]

[0234] wherein: is the quality score, is the contrast metric, is the sharpness metric, is the edge fidelity metric, is the structure fidelity metric, and are the weights of each index, and the experience values are 0.3, 0.2, 0.3 and 0.2, respectively.

[0235] If the quality score​​​​​​​ below a preset threshold (0.65 in practice), the parameters of the aforementioned processing link are adjusted, and the processing is performed again, forming a closed-loop optimization mechanism.

[0236] The innovation of the multi-feature fusion optimization unit 35 is that, through feature selection, adaptive weight calculation and quality feedback mechanism, high-quality enhancement of the blood vessel image is realized, especially in maintaining edge sharpness and structural integrity, which provides a high-quality visual basis for catheter navigation.

[0237] Referring to Figure 8 , the blood vessel three-dimensional model construction module 2 includes a three-dimensional data acquisition unit 21, a blood vessel structure identification unit 22, a blood vessel layering and division unit 23, a blood vessel connection processing unit 24 and a blood vessel tree construction unit 25.

[0238] The three-dimensional data acquisition unit 21 is used to read three-dimensional image data and perform three-dimensional spatial reconstruction. Specifically, this unit receives three-dimensional ultrasound data transmitted by the real-time image acquisition module 1, or reads pre-acquired CT / MRI data, and constructs an initial three-dimensional spatial model through voxel reconstruction technology. In transcatheter occlusion, high-resolution CT angiography (CTA) data is preferably used for reconstruction, with a resolution of 0.5-0.7mm.

[0239] The blood vessel structure identification unit 22 is connected with the three-dimensional data acquisition unit 21, and is used to obtain the blood vessel trunk and blood vessel branches, as well as the starting point and ending point of the branches. This unit uses a method based on region growing and centerline extraction to identify the blood vessel structure:

[0240] First, perform blood vessel enhancement filtering to improve the contrast between blood vessels and background;

[0241] Then extract the blood vessel region through region growing and threshold segmentation;

[0242] Finally, use a thinning algorithm to extract the blood vessel centerline to obtain the blood vessel skeleton representation.

[0243] Preferably, Frangi filter is used for blood vessel enhancement, which has good enhancement effect on tubular structures based on eigenvalue analysis of Hessian matrix.

[0244] The blood vessel layering and division unit 23 is connected with the blood vessel structure identification unit 22, and is used to divide the blood vessel trunk and blood vessel branches into blood vessel segments and blood vessel layers. This unit organizes the blood vessel network into a hierarchical structure based on the blood vessel topology: the first layer is the main blood vessel trunk (such as the aorta, pulmonary trunk, etc.); the second layer is the main branch (such as the left and right pulmonary arteries, etc.); the third layer is the secondary branch;

[0245] Each layer of blood vessels is further divided into vessel segments, which are parts of blood vessels with relatively uniform diameter and curvature, usually segmented at vessel bifurcations or where the vessel morphology changes significantly.

[0246] The vessel connection processing unit 24 is connected to the vessel layering and dividing unit 23, and is used to establish vessel segments connecting blood vessel branches and main trunks according to the starting points and ending points of the blood vessels. This unit processes the connection relationship between the blood vessel branches and the main trunks, ensuring the integrity and continuity of the model:

[0247] Identify the intersection points of the blood vessel branches and the main trunks;

[0248] Construct connected vessel segments to ensure smooth transition of the blood vessels at the intersection;

[0249] Handle complex situations where multiple blood vessels intersect to ensure correct topology.

[0250] Preferably, a spline interpolation method is used when processing the connection relationship to ensure the continuity of the curvature at the connection and avoid sharp transitions.

[0251] The vessel tree construction unit 25 is connected to the vessel connection processing unit 24, and is used to merge the connected vessel segments to form a complete vessel tree structure. This unit organizes all vessel segments into a hierarchical tree structure, facilitating navigation and visualization:

[0252] Establish a hierarchical index to record the parent-child relationship of each vessel segment;

[0253] Merge adjacent vessel segments to eliminate redundant nodes;

[0254] Optimize the tree structure to ensure that the topology is logical and consistent with anatomical characteristics.

[0255] The final generated vessel tree model not only preserves the geometric characteristics of the blood vessels, but also contains topological relationship information, providing a complete spatial reference for the navigation module.

[0256] The innovation of the vessel three-dimensional model construction module 2 lies in that through layering and dividing and connection processing, it realizes efficient expression of complex blood vessel networks, especially suitable for navigation requirements in transcatheter occlusion, and can accurately represent the spatial relationship and morphological characteristics of blood vessels.

[0257] Referring to Figure 9 , the navigation module 4 includes a registration processing unit 41, a position tracking unit 42, a trajectory prediction unit 43, a spatial mapping unit 44, and a navigation information generation unit 45.

[0258] The registration processing unit 41 is used to calculate the spatial transformation relationship between the vessel three-dimensional model and the enhanced three-dimensional spatial image of the blood vessels. This unit realizes accurate registration of the model and real-time images, and is the basis for accurate navigation:

[0259] First, extract feature points in two kinds of data (such as blood vessel bifurcation points, blood vessel curvature extreme points, etc.);

[0260] Then use the iterative closest point (ICP) algorithm for coarse registration;

[0261] Finally, through the non-rigid registration algorithm for fine registration, processing tissue deformation.

[0262] The registration transformation relationship is expressed as:

[0263] ,

[0264] Where: is the point coordinate after transformation, is the rotation matrix, is the translation vector, is the non-rigid deformation field, used to handle local deformation.

[0265] The position tracking unit 42 is connected with the registration processing unit 41, used to track the real-time position of the catheter tip in the three-dimensional space of the blood vessel. This unit processes X-ray or ultrasound images to locate the catheter tip position in real time:

[0266] First, identify the catheter tip through image processing techniques (such as high contrast, special shape markers, etc.);

[0267] Then smooth the trajectory through time series filtering (such as Kalman filtering) to reduce noise effects;

[0268] Finally, restore the two-dimensional projection position to the three-dimensional space position.

[0269] In transcatheter occlusion, the catheter tip position update frequency is usually 10-25 Hz, ensuring real-time requirements.

[0270] The trajectory prediction unit 43 is connected with the position tracking unit 42, used to predict the future motion trajectory of the catheter tip based on the historical position data of the catheter tip. This unit uses a time series prediction model to predict the future position of the catheter, providing prospective guidance for the doctor:

[0271] Establish a dynamic model of catheter motion, considering velocity, acceleration and constraint conditions;

[0272] Train the prediction model through historical trajectory data;

[0273] Generate a prediction trajectory within a future time window (such as 0.5-2 seconds).

[0274] Preferably, trajectory prediction is performed using an autoregressive model or a recurrent neural network model, with model training data coming from historical procedure records and real-time operation data.

[0275] The spatial mapping unit 44 is connected to the trajectory prediction unit 43 and the registration processing unit 41, and is used to map the catheter tip position to the corresponding position of the blood vessel three-dimensional model. This unit realizes accurate mapping of real-time position and model space:

[0276] The real-time position of the catheter tip is converted to the model space through the registration transformation relationship;

[0277] The closest distance from the catheter tip to the blood vessel centerline is calculated;

[0278] The exact position of the catheter in the blood vessel tree (the blood vessel segment and the relative position) is determined.

[0279] The mapping accuracy directly affects the navigation effect, and the preferred mapping error is controlled within 1mm, meeting the clinical needs.

[0280] The navigation information generation unit 45 is connected to the spatial mapping unit 44, and is used to generate catheter operation suggestions and path planning information. Based on the current position and the target position, this unit generates navigation guidance information:

[0281] The best path from the current position to the target position is calculated;

[0282] Directional indications, distance information and warning prompts are generated;

[0283] Operation suggestions such as steering angle, push distance, etc. are provided.

[0284] Before the release of the occluder, the system will also provide release position evaluation, analyze the matching degree of the occluder and the defect site, and assist the doctor in making decisions.

[0285] The innovation of the navigation module 4 lies in the combination of accurate registration, real-time tracking and trajectory prediction, which realizes high-precision and forward-looking catheter navigation, greatly improving the operation precision and safety of transcatheter occlusion.

[0286] Referring to Figure 10 , the fusion module 5 includes a data synchronization unit 51, a feature matching unit 52, a transparency adjustment unit 53, a view optimization unit 54 and a fusion rendering unit 55.

[0287] The data synchronization unit 51 is used to ensure the synchronization of the blood vessel three-dimensional space image after enhancement processing and the blood vessel three-dimensional model in time and space. This unit handles the time alignment and spatial coordination of different source data:

[0288] Time synchronization: align different data streams through timestamps and compensate for processing delays;

[0289] Spatial alignment: Ensure different data use the same coordinate system and scale.

[0290] In practical systems, a timestamp-based buffering mechanism is adopted to ensure data synchronization, with synchronization accuracy preferably controlled within 50ms.

[0291] The feature matching unit 52 is connected to the data synchronization unit 51 and is used to identify corresponding feature points in the two types of data. This unit establishes a correspondence between different data through a feature matching algorithm:

[0292] Extract feature points (such as blood vessel bifurcation points, anatomical landmark points, etc.) in images and models;

[0293] Use feature descriptors (such as SIFT, SURF, etc.) to represent feature points;

[0294] Establish a correspondence through a feature matching algorithm to eliminate false matches.

[0295] Preferably, feature matching is combined with domain knowledge (such as anatomical features) to improve matching accuracy.

[0296] The transparency adjustment unit 53 is connected to the feature matching unit 52 and is used to dynamically adjust the model transparency according to clinical needs and image quality. This unit balances the visibility of real-time images and three-dimensional models through intelligent transparency control:

[0297] Automatically adjust the model transparency according to the image quality score;

[0298] Reduce model transparency at key locations (such as near defect sites) to enhance visibility;

[0299] Provide a manual adjustment interface to meet the preferences of different doctors.

[0300] The transparency value ranges from 0 to 1, with 0 indicating complete transparency and 1 indicating complete opacity. The default value is usually set to 0.3-0.7 and is dynamically adjusted according to the specific scene.

[0301] The view optimization unit 54 is connected to the transparency adjustment unit 53 and is used to optimize the model display view according to the operator's line of sight and surgical needs. This unit provides the best observation angle through view calculation and adjustment:

[0302] Track the doctor's head position and line of sight direction;

[0303] Calculate the current area of interest (such as near the catheter tip);

[0304] Automatically adjust the view to ensure optimal visibility of key areas.

[0305] In transcatheter occlusion, the preferred perspective is set to an angle that can simultaneously observe the catheter tip, the current blood vessel segment, and the front blood vessel path, providing a sense of space and direction for operation.

[0306] The fusion rendering unit 55 is connected with the perspective optimization unit 54, for fusing the adjusted data into a final display image. This unit uses advanced graphics rendering technology to generate intuitive and information-rich fusion display:

[0307] A method combining voxel rendering and surface rendering is used;

[0308] An illumination model is applied to enhance depth perception;

[0309] Auxiliary information (such as distance markers, warning prompts, etc.) is added;

[0310] The rendering efficiency is optimized to ensure real-time performance (frame rate preferably maintained above 30fps).

[0311] The final rendering result is displayed to the doctor through the AR device, providing an immersive navigation experience.

[0312] The innovation of the fusion module 5 lies in the intelligent fusion and dynamic adjustment of multi-source data, which realizes information-rich, intuitive and clear visualization, effectively supporting the doctor's operation decision-making in transcatheter occlusion.

[0313] The AR device 6 includes a display component, a camera component, a processing component, an interaction component, a communication component, and a positioning component.

[0314] The display component 61 is used to display the fused blood vessel three-dimensional model in three-dimensional form. Preferably, light waveguide perspective display technology is used, and the doctor can simultaneously observe the patient and the superimposed virtual content:

[0315] Resolution: preferably 1920x1080 or higher;

[0316] Field of view: preferably more than 40°;

[0317] Refresh rate: preferably more than 60Hz, meeting the medical display requirements.

[0318] The camera component 62 is used to capture real-time scene images. A stereo camera is configured to capture three-dimensional information of the operating site:

[0319] Resolution: preferably 1080p or higher;

[0320] Frame rate: preferably 30fps or higher;

[0321] An infrared sensor is configured to enhance the imaging effect in low-light environments.

[0322] The processing component 63 is connected with the display component 61 and the camera component 62, and is used for integrating the surgical scene image and the blood vessel three-dimensional model. A high-performance processor is built-in for image processing and model rendering:

[0323] Processing capacity: supports real-time image processing and 3D rendering;

[0324] Memory: preferably 8GB or more;

[0325] Heat dissipation system: ensure stable operation during long time surgery.

[0326] The interaction component 64 is connected with the processing component 63, and is used for receiving the gesture or voice instruction of the doctor and adjusting the display content. Various interaction modes are provided to adapt to the special needs of the surgical environment:

[0327] Gesture recognition: contactless operation, keep the surgical area sterile;

[0328] Voice control: supports key instruction recognition, response rate preferably more than 95%;

[0329] Gaze tracking: automatically adjust the focus area according to the doctor's gaze.

[0330] The communication component 65 is connected with the processing component 63, and is used for data exchange with the navigation module and the fusion module. Various communication modes are supported to ensure the stability and security of data transmission:

[0331] Wireless communication: supports Wi-Fi6, Bluetooth5.0 and other high-speed wireless transmission;

[0332] Wired interface: provides a backup connection method to ensure stability in critical scenarios;

[0333] Data encryption: protect the safety of patient sensitive information.

[0334] The positioning component 66 is connected with the processing component 63, and is used for real-time acquisition of the position and attitude of the AR device in space. Multi-sensor fusion technology is adopted to achieve high-precision positioning: Inertial Measurement Unit (IMU): track the attitude and motion of the device; Optical tracking: identify environmental feature points through the camera to assist positioning; Positioning accuracy: preferably controlled within 1mm and 0.5° to meet the needs of medical navigation.

[0335] The innovation of the AR device 6 is that through multi-sensor fusion and high-performance processing, an immersive surgical navigation experience is achieved, especially in terms of device lightness, interaction convenience and display effect, which adapts to the special needs of the surgical environment.

[0336] The working process of the transcatheter occlusion enhanced reality blood vessel fusion navigation system in actual application is as follows:

[0337] 1. Preoperative preparation:

[0338] Obtain CTA or MRI image data of the patient; construct a three-dimensional model of the patient's blood vessels using the blood vessel three-dimensional model construction module 2; mark and measure the defect location, select the appropriate occluder size;

[0339] 2. System setup:

[0340] The doctor wears the AR device 6; the system automatically performs spatial calibration and registration; load the three-dimensional model of the blood vessels into the system;

[0341] 3. Catheter insertion and navigation:

[0342] The real-time image acquisition module 1 acquires ultrasound and X-ray images; the image enhancement module 3 performs enhancement processing on the images; the navigation module 4 tracks the catheter tip position in real time; the fusion module 5 fuses the enhanced images with the three-dimensional model; the AR device 6 displays the fused blood vessel model and catheter position on the corresponding position on the patient's body surface.

[0343] 4. Occluder positioning and release:

[0344] When the catheter approaches the defect location, the system automatically adjusts the display scale and transparency; the system provides accurate position feedback and operation suggestions; the doctor adjusts the occluder position according to the AR display information; the system evaluates the occluder and defect matching degree, and gives release suggestions;

[0345] 5. Postoperative evaluation:

[0346] The system displays the occlusion effect in real time; assesses whether there is residual shunt through enhanced images; records surgical data for subsequent analysis and improvement;

[0347] In a specific application case, the system is used for a case of atrial septal defect occlusion, the system successfully guides the doctor to accurately navigate the catheter to the defect location, the occluder is accurately released at one time, the occlusion effect is good, the whole operation time is reduced by 25% compared with the traditional method, the contrast agent dosage is reduced by 40%, and the radiation dose received by the patient and the doctor is also significantly reduced.

[0348] The augmented reality blood vessel fusion navigation system for transcatheter occlusion of the present application significantly improves the accuracy and safety of transcatheter occlusion through innovative image enhancement technology, accurate blood vessel modeling and intuitive AR navigation, reduces the risk of complications, shortens the operation time, and has broad clinical application prospects.

[0349] The above merely provides the preferred embodiments of the present application, and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall fall into the scope of protection of the present application.

Claims

1. An augmented reality vascular fusion navigation system for transcatheter occlusion procedures, characterized in that, The application relates to a real-time image acquisition module connected with an ultrasonic catheter, which is used for acquiring a blood vessel three-dimensional space image; a blood vessel three-dimensional model construction module connected with the real-time image acquisition module, which is used for constructing a blood vessel three-dimensional model based on the blood vessel three-dimensional space image; an image enhancement module connected with the real-time image acquisition module, which is used for performing enhancement processing on the blood vessel three-dimensional space image; the image enhancement module comprises: a space self-adaptive transformation unit, which is used for performing space self-adaptive transformation processing on the blood vessel three-dimensional space image to generate a gray image; a key point identification and target tracking unit connected with the space self-adaptive transformation unit, which is used for identifying blood vessel key points on the gray image and constructing a blood vessel connected structure; a region self-adaptive segmentation unit connected with the key point identification and target tracking unit, which is used for calculating blood vessel region information entropy and performing self-adaptive segmentation based on the blood vessel connected structure; an edge feature extraction and splicing unit connected with the region self-adaptive segmentation unit, which is used for extracting a region edge of the segmented blood vessel region and generating a region feature image; a multi-feature fusion optimization unit connected with the edge feature extraction and splicing unit, which is used for performing feature selection, weighted fusion and optimization on the region feature image; a navigation module connected with the blood vessel three-dimensional model construction module and the image enhancement module, which is used for performing fusion comparison based on the blood vessel three-dimensional model and the enhanced blood vessel three-dimensional space image to determine a catheter tip position; a fusion module connected with the image enhancement module and the blood vessel three-dimensional model construction module, which is used for fusing the enhanced blood vessel three-dimensional space image and the blood vessel three-dimensional model to generate fusion display data; and an AR device connected with the navigation module and the fusion module, which is used for superimposedly displaying the blood vessel three-dimensional model at the catheter tip position based on the catheter tip position and the fusion display data. The space self-adaptive transformation unit comprises: an image preprocessing component, which is used for performing noise reduction and equalization processing on the blood vessel three-dimensional space image; a local characteristic analysis component connected with the image preprocessing component, which is used for calculating local region statistical characteristics of the blood vessel three-dimensional space image; a transformation function construction component connected with the local characteristic analysis component, which is used for constructing a space self-adaptive transformation function based on the local region statistical characteristics; a transformation execution component connected with the transformation function construction component, which is used for applying the space self-adaptive transformation function to process the blood vessel three-dimensional space image; and a gray conversion component connected with the transformation execution component, which is used for converting the transformed blood vessel three-dimensional space image into a gray image. The key point identification and target tracking unit comprises: a feature point detection component, which is used for detecting blood vessel feature points on the gray image; a multi-scale analysis component connected with the feature point detection component, which is used for analyzing the blood vessel feature points in different scale spaces; and a connected structure construction component connected with the multi-scale analysis component, which is used for establishing a blood vessel network connected graph based on the blood vessel feature points. ​ ​ ​ ​ ​ ​ ​ ​ ​ 2. The augmented reality blood vessel fusion navigation system of claim 1, wherein, ​ ​ ​ ​ ​ ​ 3. The augmented reality blood vessel fusion navigation system of claim 1, wherein, ​ ​ ​ ​ A key point marking component, connected with the communication structure construction component, is configured to classify and mark the blood vessel feature points; A dynamic target tracking component, connected with the key point marking component, is configured to establish a tracking model to realize dynamic tracking of the blood vessel feature points.

4. The augmented reality blood vessel fusion navigation system of claim 1, wherein, The region adaptive segmentation unit comprises: An information entropy calculation component configured to calculate information entropy of each blood vessel unit in the blood vessel communication structure; A sub-region division component, connected with the information entropy calculation component, is configured to divide each blood vessel unit into a plurality of sub-regions; A region characteristic analysis component, connected with the sub-region division component, is configured to analyze the complexity of different regions based on information entropy; An adaptive threshold generation component, connected with the region characteristic analysis component, is configured to automatically calculate a segmentation threshold according to the region information entropy; A region marking component, connected with the adaptive threshold generation component, is configured to mark the segmented regions.

5. The augmented reality vascular fusion navigation system of claim 1, wherein, The edge feature extraction and splicing unit comprises: An edge detection component configured to detect blood vessel edges by using gradient and direction information; An edge enhancement component, connected with the edge detection component, is configured to perform enhancement processing on the detected blood vessel edges; A region component extraction component, connected with the edge enhancement component, is configured to segment an image into a plurality of independent region components based on blood vessel edge information; A feature image generation component, connected with the region component extraction component, is configured to generate a feature image for each region component; An edge-guided splicing component, connected with the feature image generation component, is configured to guide accurate splicing of region components by using edge information.

6. The augmented reality blood vessel fusion navigation system of claim 1, wherein, The multi-feature fusion optimization unit comprises: A feature analysis component configured to analyze the feature image of each region component; A feature screening component, connected with the feature analysis component, is configured to screen representative features based on information value evaluation; A weight calculation component, connected with the feature screening component, is configured to calculate a fusion weight according to feature importance and reliability; A weighted fusion component, connected with the weight calculation component, is configured to integrate multiple features to generate an enhanced image by using a weighted fusion strategy; A quality evaluation component, connected with the weighted fusion component, is configured to evaluate the quality of the fusion result and feed back the evaluation result to the aforementioned processing links.

7. The augmented reality vascular fusion navigation system of claim 1, wherein, The blood vessel three-dimensional model construction module comprises: A three-dimensional data acquisition unit configured to read three-dimensional image data and perform three-dimensional space stereoscopic reconstruction; A blood vessel structure identification unit, connected with the three-dimensional data acquisition unit, is configured to obtain blood vessel trunks and blood vessel branches as well as starting points and endpoints of the branches; A blood vessel layering and division unit, connected with the blood vessel structure identification unit, is configured to divide the blood vessel trunks and blood vessel branches into blood vessel segments and blood vessel layers; A blood vessel connection processing unit, connected with the blood vessel layering and division unit, is configured to establish blood vessel segments connected between blood vessel branches and blood vessel trunks according to the starting points and endpoints of the blood vessels; A blood vessel tree construction unit, connected with the blood vessel connection processing unit, is configured to merge the connected blood vessel segments to form a complete blood vessel tree structure.

8. The augmented reality blood vessel fusion navigation system of claim 1, wherein, The navigation module comprises: A registration processing unit configured to calculate the spatial transformation relationship between the blood vessel three-dimensional model and the blood vessel three-dimensional space image after the enhancement processing; a position tracking unit, connected to the registration processing unit, for tracking real-time position of the catheter tip in the three-dimensional space of the blood vessel; a trajectory prediction unit, connected to the position tracking unit, for predicting future motion trajectory of the catheter tip based on historical position data of the catheter tip; a space mapping unit, connected to the trajectory prediction unit and the registration processing unit, for mapping the position of the catheter tip to the corresponding position of the three-dimensional model of the blood vessel; a navigation information generation unit, connected to the space mapping unit, for generating catheter operation suggestions and path planning information.

9. The augmented reality vascular fusion navigation system of claim 1, wherein, The fusion module comprises: a data synchronization unit for ensuring synchronization of the three-dimensional space image of the blood vessel after enhanced processing and the three-dimensional model of the blood vessel in time and space; a feature matching unit, connected to the data synchronization unit, for identifying corresponding feature points in the two kinds of data; a transparency adjustment unit, connected to the feature matching unit, for dynamically adjusting the transparency of the model according to clinical needs and image quality; a view angle optimization unit, connected to the transparency adjustment unit, for optimizing the display view angle of the model according to the operator's line of sight and the need for surgery; a fusion rendering unit, connected to the view angle optimization unit, for rendering the adjusted data into a final display image.

10. The augmented reality vascular fusion navigation system of claim 1, wherein, The AR device comprises: a display component for displaying the fused three-dimensional model of the blood vessel in three-dimensional form; a camera component for capturing real-time scene images of the surgery; a processing component, connected to the display component and the camera component, for integrating the surgery scene images and the three-dimensional model of the blood vessel; an interaction component, connected to the processing component, for receiving the doctor's gestures or voice instructions and adjusting the display content; a communication component, connected to the processing component, for data exchange with the navigation module and the fusion module; a positioning component, connected to the processing component, for real-time acquisition of the position and attitude of the AR device in space.

Images