A method for repairing and topological analysis of a liver CT image blood vessel segmentation mask

Abstract

The application provides a kind of liver CT image blood vessel segmentation mask fracture repair and topological analysis method, specific steps include: 1) skeletonization is carried out to liver blood vessel segmentation mask, and the blood vessel skeleton, skeleton end point and blood vessel radius rough value at skeleton point are extracted;2) without considering the removal of hepatic vein and portal vein error adhesion, for the purpose of fracture repair, obtain connected blood vessel center line ①;3) considering the removal of hepatic vein and portal vein possible error adhesion, set more strict connection condition, obtain the broken blood vessel center line, then for the purpose of classification, obtain the blood vessel center line ② containing fracture and obvious distinction of hepatic vein and portal vein;4) merge the above center line ① and ②, reconstruct the missing blood vessel segment in the original blood vessel segmentation mask, merge into the original segmentation mask, obtain the final fracture repair blood vessel mask and center line topological structure.

Description

TECHNICAL FIELD

[0001] The present application relates to an image analysis technology, belonging to the technical field of computer image processing, and in particular to a liver CT image blood vessel segmentation mask fracture repair and topological analysis method. BACKGROUND

[0002] In the field of medical image analysis, the processing and analysis of liver CT images have become an important research direction, especially in the diagnosis and treatment planning of liver lesions. The blood vessel structure in the liver CT image is crucial anatomical information in diagnosis and treatment. Precise segmentation, reconstruction of blood vessels, and extraction and classification of blood vessel centerlines are of great significance for subsequent lesion detection, surgical planning, etc.

[0003] In recent years, the blood vessel segmentation technology in CT images has developed rapidly. Representative technologies include deep learning techniques based on convolutional neural networks, which have been widely used in medical image analysis. These algorithms can quickly and accurately segment blood vessels in liver CT images, but in complex blood vessel morphology, blood vessel fractures or adhesions can easily occur, and it is difficult to distinguish between hepatic veins and portal veins, which affects subsequent analysis and processing. Therefore, the mask obtained by the segmentation algorithm usually needs to be post-processed according to the application requirements. Many post-processing methods usually focus on a single task, often only focusing on the repair of blood vessel connectivity or simply classifying blood vessels, which is not sufficient to meet the needs of practical applications. SUMMARY

[0004] To solve the problems of blood vessel fracture, different blood vessel adhesion, and blood vessel mask classification in liver CT image blood vessel segmentation mask, the present application provides a liver CT image blood vessel segmentation mask fracture repair and topological analysis method, which can simultaneously obtain a fracture-repaired blood vessel mask and a centerline topological structure to meet the needs of blood vessel segmentation mask analysis.

[0005] In order to achieve the above purpose, the present application provides the following technical scheme:

[0006] A liver CT image blood vessel segmentation mask fracture repair and topological analysis method, comprising the following steps:

[0007] Step 1, performing skeletonization on the liver CT image blood vessel segmentation mask to obtain skeleton points and skeleton end points , and estimating a blood vessel radius rough value based on the maximum radial distance method

[0008] Step 2, based on the blood vessel skeleton points , skeleton end points and blood vessel radius rough value Without considering the removal of the wrong adhesions of the hepatic vein and portal vein, the centerline of the adhesion vessels is directly formed from the skeleton points through depth-first search with the purpose of fracture repair. , then Connect the fractured part and obtain the centerline of the connected blood vessel①;

[0009] Step 3: Based on the vascular skeleton points obtained in step 1 , Skeleton Endpoint The rough value of the vessel radius obtained in step 2 is considered to remove the possible erroneous adhesion of the hepatic vein and portal vein, set more stringent connection conditions, add constraints such as tangent vector angle and branch length in the depth-first search process, and form the broken vessel centerline from the skeleton point. ; To distinguish the hepatic vein from the portal vein, Classification was performed to obtain a vascular centerline that included the rupture and clearly distinguished the hepatic vein and portal vein②;

[0010] Step 4: Merge and cut the centerline ① obtained in step 2 and the centerline ② obtained in step 3 based on the vascular branch to obtain the merged vascular centerline. That is, the vascular topology and connecting lines ;

[0011] Step 5: Blood vessel segmentation mask and the connecting line obtained in step 4 , the radial heuristic method of the centerline tangent vector is used to accurately calculate the radius of the blood vessels of the two branches that should be connected , and The mean value is taken as the vessel radius of the junction , reconstruct the missing blood vessel segments in the original blood vessel segmentation mask , and Merge to original segmentation mask The final complete blood vessel mask is obtained.

[0012] Furthermore, the step 2 specifically includes the following process:

[0013] Step 21: Skeleton Endpoints According to the diameter of blood vessel Sort from large to small to get the sorted skeleton breakpoints, recorded as ;

[0014] Step 22: The first endpoint is the starting point, and the depth-first search method is used to connect the skeleton points in the 26 areas in sequence, and update 、 When no skeleton point is found within the 26-neighborhood, it means that the center line of the vascular tree Generation completed;

[0015] Step 23: Delete the centerline of the vessel tree The leaf branches with less than n branch nodes are used to obtain the corrected vascular tree centerline. ;

[0016] Step 24: Align the corrected vascular tree centerline Join the collection of adherent vascular centerlines;

[0017] Step 25: Repeat steps 22 to 24 until Empty to obtain the centerline of the adhered blood vessel ;

[0018] Step 26: Select a main vascular tree based on the number of nodes in the vascular tree. , the rest are sub-vessel trees ;

[0019] Step 27: Connection probability between the endpoints of the sub-vessel tree branches and the nodes of the main vessel tree , solve the following equation:

[0020]

[0021] in is the distance between all branch endpoints of the sub-vessel tree and all points on the main vessel tree, It is the difference in curvature between all branch endpoints of the sub-vessel tree and all points on the main vessel tree.

[0022] Calculate the connection probability between each sub-vascular tree branch endpoint and the main vascular tree node , and connect The two endpoints corresponding to the maximum value are used to update the main vessel tree and the sub-vessel tree;

[0023] Step 28: Circulate the connection probability solution process in step 27 until the sub-vessel tree is empty, and obtain the connected vessel centerline ①.

[0024] Furthermore, step 3 considers the adhesion of the hepatic vein and portal vein, including the rupture and the clear separation of the hepatic vein and portal vein by the vascular centerline ②, and specifically includes the following process:

[0025] Step 31: Skeleton Endpoints According to the diameter of blood vessel Sort from large to small to get the sorted skeleton breakpoints, recorded as ;

[0026] Step 32: The first end point is taken as the starting point, and the skeleton points in the 26-neighborhood are connected in turn using a depth-first search method, with the requirement that the cosine of the angle between the vector of the skeleton point and the vector of all points in the vessel branch in which the skeleton point is located be less than a given threshold a, and the 、 When a skeleton point in the 26-neighborhood cannot be found, it is indicated that the centerline of the vessel tree is generated;

[0027] Step 33: Delete the centerline of the vessel tree with a branch node number less than n in the leaf branch, to obtain a corrected centerline of the vessel tree ; n is 3 or 5;

[0028] Step 34: Delete the corrected centerline of the vessel tree with a branch node number less than m in the non-leaf branch and its sub-branches, and update 、 to obtain a re-corrected centerline of the vessel tree ;

[0029] Step 35: Add the re-corrected centerline of the vessel tree to the set of broken vessel centerlines;

[0030] Step 36: Repeat the above steps 32 to 35 until is empty, and finally obtain the broken vessel centerline ;

[0031] Step 37: According to the number of nodes contained in the vessel tree, select two main vessel trees 、 , which are the hepatic vein and the portal vein, and the rest are sub-vessel trees ;

[0032] Step 38: The attribution probability of the sub-vessel tree to the main vessel tree is solved by the following equation:

[0033]

[0034]

[0035] wherein is the distance between all branch end points of the sub-vessel tree and all points on the main vessel tree, and its weight is , is the difference in curvature between all branch end points of the sub-vessel tree and all points on the main vessel tree, and its weight is , C represents the direction similarity score, and the smaller the value, the more similar the directions of the two straight lines. is the angle between the directional vector of the end point of all branches of the child vessel tree and the directional vector of all points on the main vessel tree, is the angle between the connection line of the breakpoint and the branch where the breakpoint of the main vessel tree is located, is the angle between the connection line of the breakpoint and the branch where the breakpoint of the child vessel tree is located, and are weighting coefficients, used to adjust the influence degree of different angle terms on the total similarity score. Among them, 、 、 、 According to the actual situation, such as: for the hepatic vein, w1=1.5, w2=0.01, w3=0.1, w4=1.0, for the portal vein, w1=2, w2=0.1, w3=0.1, w4=1.0.

[0036] Calculate the attribution probability of the child vessel tree belonging to each main vessel tree , and connect the two end points corresponding to the maximum value in the two groups , update the corresponding main vessel tree and child vessel tree;

[0037] Step 39: Recursively calculate the attribution probability solving process in the above step 38 until the child vessel tree is empty, and obtain the vessel center line ② containing the break and the obvious distinction between the hepatic vein and the portal vein.

[0038] Further, the merging in step 4 includes the following process:

[0039] According to the center line ②, determine which category each point in each branch of the center line ① belongs to, if more than half of the points on the branch center line belong to the portal vein, then the branch belongs to the portal vein, if more than half of the points belong to the hepatic vein, then the branch belongs to the hepatic vein, if the judgment result is that the branch neither belongs to the portal vein nor belongs to the hepatic vein, then eliminate the error blood vessel segment;

[0040] The hepatic vein and the portal vein are merged into the merged blood vessel center line , obtaining the blood vessel topology and connection line .

[0041] Further, the broken blood vessel segment reconstruction and mask merging in step 5 includes the following process:

[0042] Step 51: Use the Euclidean distance transform to obtain the distance from each point in the blood vessel to the nearest boundary;

[0043] Step 52: In the vicinity of each center line point of the blood vessel, expand along the orthogonal direction of the tangent vector until the boundary of the blood vessel is reached;

[0044] Step 53: The median of the distances extended along multiple orthogonal directions is the radius of the blood vessel at that location;

[0045] Step 54: The average of the vascular radius of all points on the center line of the two connecting branches is used as the vascular radius of the connecting line. ;

[0046] Step 55: Connect the wires and the vessel radius at the connecting line ,based on The spherical structure is reconstructed using the binary expansion method to reconstruct the broken blood vessel segment. ;

[0047] Step 56: Merge and the original segmentation mask to obtain the final complete blood vessel mask.

[0048] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0049] Unlike conventional methods that only perform connectivity repair or centerline classification, the present invention sets corresponding connection conditions according to different purposes, obtains connection centerlines and classification centerlines from skeleton points respectively, then merges the connected vascular centerlines and the classified vascular centerlines, and finally reconstructs and visualizes them. It can simultaneously realize vascular centerline extraction, repair of broken vascular segments and vascular topology analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a flow chart of a method for repairing breaks and performing topological analysis of blood vessel segmentation masks in liver CT images provided by the present invention.

[0051] Figure 2 Schematic diagram of the results of a method for repairing breaks and performing topological analysis on blood vessel segmentation masks in liver CT images, provided by the present invention. (a) The original segmentation mask and its corresponding centerlines. Circles indicate broken vessels, and white centerlines represent unclassified centerlines before processing. (b) The image after applying the present method. The circled breaks have been repaired according to the algorithm, and the centerlines now distinguish between the hepatic vein (blue) and portal vein (green). DETAILED DESCRIPTION

[0052] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are intended only to illustrate the present invention and are not intended to limit the scope of the present invention. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, and the terms "inward" and "outward" refer to directions toward or away from the geometric center of a particular component, respectively.

[0053] This embodiment provides a method for repairing the breakage and topological analysis of the blood vessel segmentation mask in the liver CT image. The process is as follows: Figure 1 As shown, the following steps are included:

[0054] Step 101: segment the blood vessels in the liver CT image. Perform skeletonization to obtain skeleton points and skeleton endpoints , and estimate the rough value of the blood vessel radius based on the maximum radial distance method .

[0055] In this step, the input is the segmentation mask of the blood vessels in the liver CT image based on deep learning, and the skeleton extraction method based on Euler features is used to extract the blood vessel skeleton points. and skeleton endpoints , where the number of loops is set to 20. Then, based on the vascular segmentation mask, the vascular boundary is calculated. Then, the distance from the vascular skeleton node to the nearest non-zero coordinate point on the vascular boundary is calculated on the XY, YZ, and XZ planes respectively. 、 、 Finally, 、 、 The maximum value is taken as the rough value of the blood vessel radius at the skeleton point .

[0056] Step 102: Based on the blood vessel skeleton points obtained in step 101 , Skeleton Endpoint and vessel radius Without considering the removal of the wrong adhesions of the hepatic vein and portal vein, the centerline of the adhesion vessels is directly formed from the skeleton points through depth-first search with the purpose of fracture repair. , then Connect the broken part and obtain the center line of the connected blood vessel①.

[0057] In this step, the specific implementation plan is as follows:

[0058] 1) Skeleton endpoints According to the diameter of blood vessel Sort from large to small to get the sorted skeleton breakpoints, recorded as ;

[0059] 2) The first endpoint is the starting point, and the depth-first search method is used to connect the skeleton points in the 26 areas in sequence, and update 、 When no skeleton point is found within the 26-neighborhood, it means that the center line of the vascular tree Generation completed;

[0060] 3) Delete the centerline of the vascular tree The leaf branches with less than n branch nodes are used to obtain the corrected vascular tree centerline. ;

[0061] 4) The corrected vascular tree centerline Join the collection of adherent vascular centerlines;

[0062] 5) Repeat steps 2) to 4) until Empty to obtain the centerline of the adhered blood vessel ;

[0063] 6) Select a main vascular tree based on the number of nodes contained in the vascular tree , the rest are sub-vessel trees ;

[0064] 7) Connection probability between the endpoints of the sub-vascular tree branches and the nodes of the main vascular tree , solve the following equation:

[0065]

[0066] in is the distance between all branch endpoints of the sub-vessel tree and all points on the main vessel tree, It is the difference in curvature between all branch endpoints of the sub-vessel tree and all points on the main vessel tree.

[0067] Calculate the connection probability between each sub-vascular tree branch endpoint and the main vascular tree node , and connect The two endpoints corresponding to the maximum value are used to update the main vessel tree and the sub-vessel tree;

[0068] 8) The connection probability solution process in step 7) is repeated until the sub-vessel tree is empty, and the connected vessel centerline ① is obtained.

[0069] Step 103: Based on the blood vessel skeleton points obtained in step 101 , Skeleton Endpoint and the rough value of the blood vessel radius obtained in step 102 , considering removing possible erroneous adhesions between the hepatic vein and portal vein, setting stricter connection conditions, adding constraints such as tangent vector angle and branch length in the depth-first search process, and forming the broken vascular centerline from the skeleton points ; To distinguish the hepatic vein from the portal vein, Classification was performed to obtain the vascular centerline containing the rupture and with the hepatic vein and portal vein clearly distinguished②.

[0070] In this step, the specific implementation plan is as follows:

[0071] 1) Skeleton endpoints According to the diameter of blood vessel Sort from large to small to get the sorted skeleton breakpoints, recorded as ;

[0072] 2) The first endpoint is the starting point, and the depth-first search method is used to connect the skeleton points in the 26 areas in sequence. It is required that the cosine value of the angle between the vector between the skeleton point and the connection point and the vector of all points in the vascular branch where the skeleton point is located is less than a given threshold a, and update 、 When no skeleton point is found within the 26-neighborhood, it means that the center line of the vascular tree Generation completed;

[0073] 3) Delete the centerline of the vascular tree The leaf branches with less than n branch nodes in the middle obtain the corrected vascular tree centerline ;

[0074] 4) Delete the corrected vascular tree centerline The non-leaf branches and their sub-branches whose number of branch nodes is less than m, and update 、 , obtain the corrected vascular tree centerline ;

[0075] 5) The corrected vascular tree centerline Join the broken vessel centerline collection;

[0076] 6) Repeat the above steps 2) to 5) until Empty, finally get the center line of the broken blood vessel ;

[0077] 7) Select two main vascular trees based on the number of nodes contained in the vascular tree 、 , respectively the hepatic vein and portal vein, the rest are sub-vascular trees ;

[0078] 8) The probability that a sub-vessel tree belongs to the main vessel tree , solve the following equation:

[0079]

[0080]

[0081] in is the distance between all branch endpoints of the sub-vascular tree and all points on the main vascular tree, and its weight is recorded as , It is the difference in curvature between all branch endpoints of the sub-vessel tree and all points on the main vessel tree, and its weight is recorded as , C represents the direction similarity score. The smaller the value, the more similar the directions of the two lines are. It is the angle between the direction vectors of all branch endpoints of the sub-vessel tree and the direction vectors of all points on the main vessel tree. It is the angle between the breakpoint connecting line and the branch of the main vascular tree where the breakpoint is located. is the angle between the breakpoint connecting line and the branch where the breakpoint of the sub-vascular tree is located. and is a weighting coefficient used to adjust the influence of different angle items on the overall similarity score. w1, w2, w3, and w4 are determined based on the actual situation. For example, for the hepatic vein, w1=1.5, w2=0.01, w3=0.1, and w4=1.0; for the portal vein, w1=2, w2=0.1, w3=0.1, and w4=1.0.

[0082] Calculate the probability that the sub-vessel tree belongs to each main vessel tree , and connect the two groups The two endpoints corresponding to the maximum value are used to update the corresponding main vessel tree and sub-vessel tree;

[0083] 9) Repeat the calculation process of the attribution probability solution in step 8) above until the sub-vessel tree is empty, and obtain the vessel centerline ② that contains the break and clearly distinguishes the hepatic vein and portal vein.

[0084] Step 104: Merge and cut the centerline ① obtained in step 102 and the centerline ② obtained in step 103 based on the vascular branch to obtain the merged vascular centerline. That is, the vascular topology and connecting lines .

[0085] In this step, the specific implementation plan is as follows:

[0086] Based on centerline ②, determine which category each point of each branch in centerline ① belongs to. If more than half of the points on a branch centerline belong to the portal vein, then the branch belongs to the portal vein. If more than half of the points on the centerline of a branch belong to the hepatic vein, then the branch belongs to the hepatic vein. If the result shows that the branch belongs to neither the portal vein nor the hepatic vein, then eliminate the incorrect vessel segment.

[0087] The hepatic vein and portal vein merge into the combined vascular centerline , get the vascular topology and connection lines .

[0088] Step 105: Segmentation based on blood vessel mask and the connecting line obtained in step 104 , the radial heuristic method of the centerline tangent vector is used to accurately calculate the radius of the blood vessels of the two branches that should be connected , and The mean value is taken as the vessel radius of the junction , reconstruct the missing blood vessel segments in the original blood vessel segmentation mask , and Merge to original segmentation mask The final complete blood vessel mask is obtained.

[0089] In this step, the specific implementation plan is as follows:

[0090] 1) Use the Euclidean distance transform to obtain the distance from each point in the blood vessel to the nearest boundary;

[0091] 2) Near each centerline point of the blood vessel, expand along the orthogonal direction of the tangent vector until reaching the boundary of the blood vessel;

[0092] 3) The median of the distances extended along multiple orthogonal directions is the vessel radius at that location;

[0093] 4) The average of the vascular radius at all points on the center line of the two connecting branches is taken as the vascular radius at the connecting line. ;

[0094] 5) According to the connection line and the vessel radius at the connecting line ,based on The spherical structure of the original liver CT size was used to reconstruct the broken vessel segments using the binary expansion method. ;

[0095] 6) Use logical OR operation to merge the vessel segmentation masks and vascular segments , and obtain the final complete blood vessel mask ;

[0096] 7) Finally, the merged vascular mask is reconstructed based on the application of the surface reconstruction algorithm; using geometric reconstruction, multiple discrete points are connected to form a continuous line to reconstruct the vascular topology.

[0097] like Figure 2 As shown in the figure, (a) the original segmentation mask and the corresponding centerline, the circle represents the broken blood vessel, and the white centerline is the unclassified centerline before processing; (b) is the image after the method of the present invention is applied, the break at the circle has been repaired according to the algorithm, and the centerline distinguishes the hepatic vein (blue) and the portal vein (green).

[0098] The technical means disclosed in the present application are not limited to the technical means disclosed in the above-mentioned embodiments, and include technical solutions composed of any combination of the above technical features.

Claims

1. A method for repairing the breakage and topological analysis of blood vessel segmentation masks in liver CT images, characterized in that: The steps include: Step 1: Segment the blood vessels in the liver CT image using the mask S mask Perform skeletonization to obtain skeleton points P skel and the skeleton endpoint P skel_end , and estimate the rough value of the blood vessel radius R based on the maximum radial distance method rough ; Step 2: Based on the blood vessel skeleton point P obtained in step 1 skel , skeleton endpoint P skel_end and the rough value of the vessel radius R rough Without considering the removal of the wrong adhesions of the hepatic vein and portal vein, the center line C of the adhesion is directly formed from the skeleton points through depth-first search with the purpose of fracture repair. adhesion , then C adhesion Connect the fractured part and obtain the centerline of the connected blood vessel①; Step 3: Based on the blood vessel skeleton point P obtained in step 1 skel , skeleton endpoint P skel_end and the rough value of the blood vessel radius R obtained in step 2 rough Consider removing possible erroneous adhesions between the hepatic vein and portal vein, set stricter connection conditions, add constraints such as tangent vector angle and branch length in the depth-first search process, and form the broken vascular centerline C from the skeleton point. noadhesion ; To distinguish the hepatic vein from the portal vein, C noadhesion Classification was performed to obtain a vascular centerline that included the rupture and clearly distinguished the hepatic vein and portal vein②; Step 4: Merge and segment the centerline ① obtained in step 2 and the centerline ② obtained in step 3 based on the vascular branch to obtain the merged vascular centerline topology C. merge and connecting line C connect ; Step 5: Based on the blood vessel segmentation mask S mask and the connecting line C obtained in step 4 connect , the radial heuristic method of the centerline tangent vector is used to accurately calculate the radius of the blood vessels of the two branches that should be connected R exact , and R exact The mean value is taken as the vessel radius R at the junction connect , reconstruct the missing blood vessel segment S in the original blood vessel segmentation mask connect , and S connect Merge into the original segmentation mask S mask The final complete blood vessel mask is obtained.

2. The method for repairing the breakage and topological analysis of the blood vessel segmentation mask in the liver CT image according to claim 1, characterized in that: The step 2 specifically includes the following process: Step 21: Skeleton Endpoint P skel_end According to the rough value of blood vessel radius R rough Sort from large to small to get the sorted skeleton breakpoints, denoted as P skel_end_sorted ; Step 22: P skel_end_sorted The first endpoint is the starting point, and the depth-first search method is used to connect the skeleton points in the 26 areas in sequence, and update P skel 、P skel_end_sorted When no skeleton point is found within the 26-neighborhood, it means that the center line C of the vascular tree tree Generation completed; Step 23: Delete the center line C of the vessel tree tree The leaf branches with less than n branch nodes are used to obtain the corrected vascular tree centerline C. tree_correct , n is 3 or 5; Step 24: The corrected vascular tree centerline C tree_correct Join the collection of adherent vascular centerlines; Step 25: Repeat steps 22 to 24 until P skel_end_sorted Empty, get the center line C of the adhered blood vessel adhesion ; Step 26: Select a main vascular tree T according to the number of nodes contained in the vascular tree main , the rest are sub-vessel trees T sub ; Step 27: Connection probability P between the endpoints of the sub-vessel tree branches and the nodes of the main vessel tree connect , solve the following equation: P connect =1-(dis+curdif) Where dis is the distance between all branch endpoints of the sub-vascular tree and all points on the main vascular tree, and curdif is the difference in curvature between all branch endpoints of the sub-vascular tree and all points on the main vascular tree; Calculate the connection probability P between each branch endpoint of the sub-vascular tree and the main vascular tree node connect , and connect P connect The two endpoints corresponding to the maximum value are used to update the main vessel tree and the sub-vessel tree; 28: The connection probability solution process in step 27 is calculated repeatedly until the sub-vessel tree is empty, and the connected vessel centerline ① is obtained.

3. The method for repairing the breakage and topological analysis of the blood vessel segmentation mask in the liver CT image according to claim 1, characterized in that: Step 3 considers the adhesion of the hepatic vein and portal vein, including the rupture and the clear distinction of the vascular centerline ② between the hepatic vein and portal vein, and specifically includes the following process: Step 31: Skeleton Endpoint P skel_end According to the rough value of blood vessel radius R rough Sort from large to small to get the sorted skeleton breakpoints, denoted as P skel_end_sorted ; Step 32: P skel_end_sorted The first endpoint is taken as the starting point, and the depth-first search method is used to connect the skeleton points in the 26 domains in sequence. It is required that the cosine value of the angle between the vector between the skeleton point and the connection point and the vector of all points in the vascular branch where the skeleton point is located is less than a given threshold a, and P is updated. skel 、P skel_end_sorted When no skeleton point is found within the 26-neighborhood, it means that the center line C of the vascular tree tree Generation completed; Step 33: Delete the center line C of the vessel tree tree The leaf branches with less than n branch nodes are used to obtain the corrected vascular tree centerline C tree_correct ; Step 34: Delete the corrected vascular tree centerline C tree_correct The non-leaf branches and their sub-branches whose number of branch nodes is less than m, and update P skel 、P skel_end_sorted , obtain the corrected vascular tree centerline C tree_rcorrect ; Step 35: The corrected vascular tree centerline C tree_rcorrect Join the broken vessel centerline collection; Step 36: Repeat steps 32 to 35 until P skel_end_sorted Empty, finally get the broken blood vessel centerline C noadhesion ; Step 37: Select two main vascular trees T according to the number of nodes contained in the vascular tree main1 、T main2 , respectively the hepatic vein and portal vein, the rest are sub-vascular trees T sub ; Step 38: The probability P of the sub-vessel tree belonging to the main vessel tree belong , solve the following equation: P belong =1-(w1×dis+w2×curdif+C) C=w3×(1-cosα)+w4×(1-cos(β+γ)) Where dis is the distance between all branch endpoints of the sub-vessel tree and all points on the main vascular tree, and its weight is w1. Curdif is the difference in curvature between all branch endpoints of the sub-vessel tree and all points on the main vascular tree, and its weight is w2. C represents the directional similarity score, and the smaller the value, the more similar the directions of the two lines are. α is the angle between the direction vectors of all branch endpoints of the sub-vessel tree and the direction vectors of all points on the main vascular tree. β is the angle between the breakpoint connecting line and the branch where the breakpoint of the main vascular tree is located. γ is the angle between the breakpoint connecting line and the branch where the breakpoint of the sub-vessel tree is located. w3 and w4 are weighting coefficients used to adjust the influence of different angle items on the total similarity score. For the hepatic vein, w1 = 1.5, w2 = 0.01, w3 = 0.1, w4 = 1.

0. For the portal vein, w1 = 2, w2 = 0.1, w3 = 0.1, w4 = 1.

0. Calculate the probability P of the sub-vessel tree belonging to each main vessel tree belong , and connect two groups of P belong The two endpoints corresponding to the maximum value are used to update the corresponding main vessel tree and sub-vessel tree; Step 39: The attribution probability solution process in step 38 is cyclically calculated until the sub-vessel tree is empty, and the vessel centerline ② containing the fracture and clearly distinguishing the hepatic vein and portal vein is obtained.

4. The method for repairing the breakage and topological analysis of the blood vessel segmentation mask in the liver CT image according to claim 1, characterized in that: The merging method in step 4 is as follows: Based on centerline ②, determine which category each point of each branch in centerline ① belongs to. If more than half of the points on a branch centerline belong to the portal vein, then the branch belongs to the portal vein. If more than half of the points on the centerline of a branch belong to the hepatic vein, then the branch belongs to the hepatic vein. If the result shows that the branch belongs to neither the portal vein nor the hepatic vein, then eliminate the incorrect vessel segment. The hepatic vein and portal vein merge into the combined vascular centerline C merge , get the vascular topology and connecting line C connect .

5. The method for repairing the breakage and topological analysis of the blood vessel segmentation mask in the liver CT image according to claim 1, characterized in that: The step 5 specifically includes the following process: Step 51: Use Euclidean distance transformation to obtain the distance from each point in the blood vessel to the nearest boundary; Step 52: near each centerline point of the blood vessel, expand along the orthogonal direction of the tangent vector until reaching the boundary of the blood vessel; Step 53: The median of the distances extended along multiple orthogonal directions is the radius of the blood vessel at that location; Step 54: The average value of the vascular radius of all points on the center line of the two connecting branches is used as the vascular radius R at the connecting line. connect ; Step 55: Connect Line C connect and the blood vessel radius R at the connecting line connect , based on R connect The spherical structure is reconstructed using the binary expansion method to obtain the broken blood vessel segment S connect ; Step 56: Merge S connect and the original segmentation mask to obtain the final complete blood vessel mask.

Images