CN107067409A - A blood vessel separation method and system - Google Patents

Abstract

The embodiment of the invention provides a blood vessel separation method and a blood vessel separation system. On one hand, in the embodiment of the present invention, first, real-time tracking path information of an original image of a blood vessel is obtained through tracking, and first topological analysis is performed on the real-time tracking path information to generate an initial segmentation result of the blood vessel, then, a blood vessel center line is extracted according to the initial segmentation result, and second topological analysis is performed on the blood vessel topological information based on the blood vessel center line according to a preset algorithm to generate a final segmentation result of the blood vessel. Therefore, the technical scheme provided by the embodiment of the invention separates the blood vessels in a two-step separation mode, and effectively improves the separation speed and the separation efficiency of the blood vessels.

Description

A blood vessel separation method and system

【Technical field】

The invention relates to the technical field of digital medical treatment, in particular to a blood vessel separation method and system.

【Background technique】

With the continuous development of surgical operations, precision surgery has put forward higher requirements for computer medical imaging systems. In hepatobiliary surgery, doctors need to carry out detailed analysis of the lesion area of the liver and the relationship with the blood vessels before the operation. To understanding. With the help of computer 3D imaging system and liver segmentation technology, doctors can have a clear, intuitive and three-dimensional grasp of the relationship between various tissues in the liver before operation, and can judge whether the operation can be carried out according to the results of automatic liver segmentation. This allows doctors to make detailed planning before the operation, and accurately removes cancerous tissue during the operation to reduce damage to healthy tissue.

During the liver segmentation process, the precise segmentation of the hepatic portal vein and the hepatic vein is one of the reference values for liver segmentation. However, in the hepatic venous phase and delayed phase scans, the contrast of the hepatic portal vein and hepatic vein is low and the noise is large. In addition to the influence of the volume effect, the relatively close hepatic portal vein and hepatic vein segment are in the process of separation. Adhesions may develop, causing parts or even all of the vessels of one vascular system to be recognized as the other.

【Content of invention】

In view of this, embodiments of the present invention provide a blood vessel separation method and system to solve the problem of low blood vessel segmentation accuracy and efficiency in the prior art.

On the one hand, an embodiment of the present invention provides a blood vessel separation method, the method comprising:

Obtain real-time tracking path information of the original vascular image by tracking;

performing a first topology analysis on the real-time tracking path information to generate an initial segmentation result of blood vessels;

extracting a blood vessel centerline according to the initial segmentation result;

According to a preset algorithm, a second topology analysis is performed on the vessel topology information based on the vessel centerline to generate a final vessel segmentation result.

According to the above-mentioned aspect and any possible implementation, an implementation is further provided. According to a preset algorithm, a second topology analysis is performed on the vascular topology information based on the vascular centerline to generate the final segmentation result of the blood vessel. Specifically include:

calculating the parameter information of each vessel centerline corresponding to the vessel segment, and storing the vessel topology based on the vessel centerline in an undirected graph;

After the connected path is searched in the undirected graph according to the preset marker points, a second topology analysis is performed on the connected path through a preset algorithm based on the parameter information to generate a final segmentation result of the blood vessel.

According to the above aspect and any possible implementation, an implementation is further provided, the parameter information includes: the average radius of the blood vessel segment, the average tangent direction of the blood vessel segment, the radius at both ends of the blood vessel segment, and the tangent at both ends of the blood vessel segment One or more of the directions.

According to the foregoing aspect and any possible implementation manner, an implementation manner is further provided, wherein the first topology analysis and the second topology analysis include a vessel path analysis and a vessel bifurcation analysis.

According to the above aspects and any possible implementation, an implementation is further provided, the method further includes:

The corresponding relationship between each blood vessel pixel point of the blood vessel and the blood vessel centerline is respectively stored.

According to the above aspect and any possible implementation, an implementation is further provided, the tracking adopts one of the ridge tracking method, the least cost path method, the blood vessel cross-section model tracking method and the tubular structure model tracking method or more.

On the other hand, an embodiment of the present invention provides a blood vessel separation system, the system comprising:

An acquisition unit, configured to acquire real-time tracking path information of the original blood vessel image through tracking;

The first processing unit is configured to perform a first topology analysis on the real-time tracking path information to generate an initial segmentation result of blood vessels;

an extracting unit, configured to extract a blood vessel centerline according to the initial segmentation result;

The second processing unit is configured to perform a second topology analysis on the vessel topology information based on the vessel centerline according to a preset algorithm to generate a final segmentation result of the vessel.

According to the foregoing aspect and any possible implementation manner, an implementation manner is further provided, the second processing unit is specifically configured to:

calculating the parameter information of each vessel centerline corresponding to the vessel segment, and storing the vessel topology based on the vessel centerline in an undirected graph;

After the connected path is searched in the undirected graph according to the preset marker points, a second topology analysis is performed on the connected path through a preset algorithm based on the parameter information to generate a final segmentation result of the blood vessel.

According to the foregoing aspect and any possible implementation manner, an implementation manner is further provided, wherein the first topology analysis and the second topology analysis include a vessel path analysis and a vessel bifurcation analysis.

According to the above aspect and any possible implementation, an implementation is further provided, the system further includes:

The storage unit is configured to respectively store the corresponding relationship between each blood vessel pixel point of the blood vessel and the blood vessel centerline.

One of the above technical solutions has the following beneficial effects:

In the embodiment of the present invention, during the initial segmentation process, the extremely obvious vascular adhesion points are separated by analyzing the local flutter information of the blood vessels. After the initial segmentation is completed, the path information and bifurcation information are obtained by analyzing the global topology information of the blood vessels. For further separation of vascular adhesion points, the embodiment of the present invention effectively improves the separation speed and efficiency of blood vessels through two-step separation.

【Description of drawings】

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. Those of ordinary skill in the art can also obtain other drawings based on these drawings without paying creative labor.

Fig. 1 is a schematic flow chart of a blood vessel separation method provided by an embodiment of the present invention;

Fig. 2 is the image of blood vessels before separation according to the embodiment of the present invention;

Fig. 3 is the blood vessel image after separation according to the embodiment of the present invention;

Fig. 4 is a functional block diagram of the blood vessel separation system provided by the embodiment of the present invention.

【detailed description】

In order to better understand the technical solutions of the present invention, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.

It should be clear that the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

Terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. As used in the embodiments of the present invention and the appended claims, the singular forms "a", "said" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise.

It should be understood that the term "and/or" used herein is only an association relationship describing associated objects, which means that there may be three relationships, for example, A and/or B, which may mean that A exists alone, and A and B exist simultaneously. B, there are three situations of B alone. In addition, the character "/" in this article generally indicates that the contextual objects are an "or" relationship.

Depending on the context, the word "if" as used herein may be interpreted as "at" or "when" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrases "if determined" or "if detected (the stated condition or event)" could be interpreted as "when determined" or "in response to the determination" or "when detected (the stated condition or event) )" or "in response to detection of (a stated condition or event)".

Please refer to FIG. 1, which is a schematic flow chart of a blood vessel separation method provided by an embodiment of the present invention. As shown in FIG. 1, the method includes the following steps:

Step S101, acquiring real-time tracking path information of the original blood vessel image through tracking.

Specifically, in the embodiment of the present invention, the real-time tracking path information of the original image of blood vessels is obtained through the blood vessel tracking algorithm, and the principle of the blood vessel tracking algorithm is specifically as follows:

After selecting the tracking starting point based on the original image of the blood vessel, determine the vessel radius, coordinates and direction of the vessel segment where the tracking starting point is located, establish the initial vessel model, and simulate the initial vessel model with the adjacent area located at the tracking starting point Match and match, optimize the initial vessel model, and then further determine the real-time tracking path information such as vessel radius, vessel direction, vessel contrast, vessel matching degree and other real-time tracking path information of the vessel segment where the tracking starting point is located after model optimization; then move the vessel model forward , using the same method as above to determine the real-time tracking path information at the next blood vessel position.

Step S102, performing a first topology analysis on the real-time tracking path information to generate an initial segmentation result of blood vessels.

In the embodiment of the present invention, the first topology analysis is performed based on analyzing the local topology information of the original blood vessel image. Among them, the analysis of local topological information can be understood as starting from a tracking point and tracking forward several layers of new vessel segments, analyzing the vessel path information and vessel bifurcation information of these layers of vessel segments, and obtaining the local Deep detection tree. Wherein, the tracking method for obtaining the first topology analysis includes multiple methods such as ridge line tracking method, minimum cost path method, blood vessel cross-section model tracking method, and tubular model tracking method. The real-time tracking path information of the original blood vessel image can be obtained by using one or more of the above tracking methods in combination. The ridge tracking method obtains a filter response image in which the pixel value in the center area of the vessel diameter in the image is increased and the pixel value in the border area of the vessel is decreased by preprocessing the image such as smoothing and filtering, and the image is obtained from an initial The point starts tracking the vessel centerline along the ridge direction of the filtered response. The Minimal Cost Path method (Minimal Cost Path) filters or calculates the energy of the image, assigns a cost value between adjacent pixels, and searches for the minimum path step by step from an initial point on the cost map through the dynamic programming method, so as to track the obtained Vascular centerline. The blood vessel cross-section model tracking method usually constructs a 2D blood vessel cross-section model. Starting from the starting point, transform the parameters of the model such as radius, contrast, and normal vector, and perform similarity matching with the blood vessel image to find the most similar position. A good model center point is used as the tracked blood vessel center point, and then move forward and repeat the above steps. The tubular structure model tracking method constructs a 3D tubular model, usually in the shape of an ellipsoid or a cylinder. Starting from the starting point, transform the parameters of the model such as radius, contrast, and direction of the tube diameter, and perform similarity matching with the blood vessel image to find the most similar , take the center point of the matched model as the center point of the tracked blood vessel, and then move forward and repeat the above steps.

Now take the tubular structure model tracking method as an example. In the tubular structure model tracking method, each tubular model represents each node, and a series of nodes with discontinuous center points are formed through tubular structure tracking. A directed graph, wherein the nodes of each tubular model in the tracing path are regarded as vertices in the directed graph, and the connection relationship between adjacent nodes in the tracing path is regarded as the edges in the directed graph. The directed graph is an acyclic tree structure, that is, a local depth detection tree. Each node of the tree diagram at least stores the vessel radius, vessel direction, vessel contrast and vessel matching degree of the vessel segment where the node is located. The connection of at least stores the blood vessel direction deviation information between the parent node and the child node and the angle information between the bifurcation child nodes.

It should be noted that the first topology analysis is used to determine obvious vascular adhesion points, for example, the following strategies can be used to determine obvious vascular adhesion points:

1. Fork Analysis

Based on the local topological information, the angle between the main branch and the sub-branch of each bifurcation is analyzed. When the direction deviation angle of the sub-vascular segment of the two bifurcations is greater than 160 degrees, it is determined as a vascular adhesion point, and the path tracing at the adhesion point is interrupted; When the direction deflection angle of the two branched sub-vessel segments is greater than 130 degrees, and the matching degree of the two sub-vessel segments at the bifurcation is greater than that of the parent vessel segment (because the parent vessel segment is at the adhesion point, the contrast and matching degree of the parent vessel segment are both biased). low), it is determined as an obvious vascular adhesion point, and the path tracing at the adhesion is interrupted.

2. Partial path analysis

Based on the local depth detection tree, calculate the change rate of the radius, angle and matching degree corresponding to each path from the root node to the leaf node, and compare it with the preset value, and determine the path that does not meet the preset conditions It is an obvious vascular adhesion point, and the path tracing at the adhesion is interrupted.

After the above steps, the first topology analysis (including bifurcation information and local path information) is performed on the real-time tracking path information, and an initial segmentation result of blood vessels (ie, an image mask) is generated.

Step S103, extracting the blood vessel centerline according to the initial segmentation result.

In the embodiment of the present invention, image thinning is performed on the initial segmentation result (i.e. image mask), which is also called skeletonization, and the center line composed of continuous pixels is quickly extracted through an extraction algorithm, wherein the extraction algorithm can be divided into categories according to categories There are two types of extraction based on morphology and extraction based on level set. Specifically, the principle of the extraction algorithm is as follows: first, the surface of the object on the image mask is corroded symmetrically, and the edge pixels that do not satisfy Euler invariance and non-simple points are corroded. Extract the centerline of each blood vessel in the initial segmentation result. Since the extraction algorithm is a technical means commonly used by those skilled in the art, it will not be described in detail here. After the blood vessel centerline is extracted, the corresponding relationship between each blood vessel pixel point of the blood vessel and the blood vessel center line is respectively stored, and then when a certain blood vessel pixel point is selected, the blood vessel center line can be directly located.

In step S104, according to a preset algorithm, a second topology analysis is performed on the vessel topology information based on the vessel centerline to generate a final vessel segmentation result.

The embodiment of the present invention further refines step S104, specifically including:

a) calculating the parameter information of each vessel centerline corresponding to the vessel segment, and storing the vessel topology based on the vessel centerline in an undirected graph;

In the embodiment of the present invention, the second topology analysis is performed based on the second topology. The second topology is to use the bifurcation point and the end point as vertices after analyzing the center line to extract the bifurcation point and the end point. The adjacency relationship is used as an edge to construct an undirected graph. Wherein, each edge of the undirected graph stores the average vessel radius and vessel direction of the vessel segment, and each vertex of the undirected graph stores the vessel radius and vessel direction at the position of the node.

b) After searching for a connected path in the undirected graph according to the preset marker points, performing a second topology analysis on the connected path through a preset algorithm based on the parameter information to generate a final segmentation result of the blood vessel.

In the embodiment of the present invention, the parameter information of the blood vessel segment that needs to be calculated includes the average radius of the blood vessel segment, the average tangent direction of the blood vessel segment, the radius of both ends of the blood vessel segment, and the tangent direction of the two ends of the blood vessel segment, and after the calculation is completed, each blood vessel segment Store the parameter information of the vascular center line, and store the vascular topology based on the vascular centerline in an undirected graph; then, after the user adds a preset marker point, search for a connected path in the undirected graph through a depth-first search algorithm, and then Connected paths are subjected to a second topology analysis. Specifically, after the user sets the preset markers, the algorithm searches out all connected paths of each marker pair, and then analyzes the parameter information stored on these connected paths.

It should be noted that the second topology analysis includes the following three steps:

1) First, find out the positions of the minimum radius, maximum front-back angle, and maximum front-back radius difference in the connected path corresponding to the preset marker points, and determine these edges in the undirected graph as suspected adhesion points.

2) Then, judge whether there is a bifurcation in each suspected adhesion. If there is no bifurcation in the suspected adhesion, then exclude the suspected adhesion; For example, the following strategies can be used to analyze the bifurcation of suspected adhesions:

a) Determine whether the average direction angle between the two bifurcated sides is greater than 130 degrees, if so, then judge it as a suspected adhesion, if not, then exclude the suspected adhesion;

b) Determine whether the average direction angle of the two bifurcated sides is greater than 90 degrees and the direction angle of the node at the bifurcation is greater than 130 degrees. If so, it is judged to be a suspected adhesion, and if not, the suspected adhesion is excluded.

3) Finally, determine the final suspected adhesion. If there is only one suspected adhesion left on each connected path after the screening of steps 1) and 2), then take this suspected adhesion as the final suspected adhesion; if there is still If there are multiple suspected adhesions, after performing adhesion similarity probability analysis on the multiple suspected adhesions, the suspected adhesion with the highest probability is selected as the final suspected adhesion.

For example, the probabilistic analysis of the similarity of the adhesion can be calculated by weighting the radius, angle deviation, direction and other parameter information on the connected path as the probability of the suspected adhesion, and then selecting the highest probability among the suspected adhesions as the connection The final suspected glue point of the path.

Fig. 2 is an image of blood vessels before separation according to the embodiment of the present invention, and Fig. 3 is an image of blood vessels after separation according to the embodiment of the present invention. Please refer to Figure 2. After the user calibrates a certain number of hepatic vein points and portal vein points in the blood vessel image before separation, the whole blood vessel image is automatically separated into a blood vessel image composed of hepatic vein and portal vein through the separation method of the embodiment of the present invention. This results in the separation of blood vessels.

The technical solutions of the embodiments of the present invention have the following beneficial effects:

In the embodiment of the present invention, during the initial segmentation process, the extremely obvious vascular adhesion points are separated by analyzing the local flutter information of the blood vessels. After the initial segmentation is completed, the path information and bifurcation information are obtained by analyzing the global topology information of the blood vessels. For further separation of vascular adhesion points, the embodiment of the present invention effectively improves the separation speed and efficiency of blood vessels through two-step separation.

The technical solution of the present invention will be further described below in conjunction with specific application scenarios.

The embodiment of the present invention combines the characteristics of the segmentation results of the hepatic portal vein and the hepatic vein, that is, the number of blood vessels is not many, the fine branches are not many, and the adhesions are usually 1-5 places, and a method for quickly identifying the adhesions of the two vascular systems and performing method of separation. During the initial segmentation process, the extremely obvious vascular adhesion points can be separated by analyzing the local flutter information of the blood vessels. For separation, the embodiment of the present invention effectively improves the separation speed and efficiency of blood vessels in the liver analysis workflow through a two-step separation method.

It can be understood that the technical solutions of the embodiments of the present invention can also be applied to other application scenarios with relatively simple vascular topology, such as the separation of cardiovascular arteries and veins; for scenarios with more complex topologies, such as separation of pulmonary arteries and veins, this The technical solutions of the embodiments of the invention can be used as auxiliary processing means.

Please refer to FIG. 4, which is a functional block diagram of a blood vessel separation system provided by an embodiment of the present invention. As shown in FIG. 4, the system includes:

An acquisition unit 410, configured to acquire real-time tracking path information of the original blood vessel image through tracking;

The first processing unit 420 is configured to perform a first topology analysis on the real-time tracking path information to generate an initial segmentation result of blood vessels;

An extraction unit 430, configured to extract a blood vessel centerline according to the initial segmentation result;

The second processing unit 440 is configured to perform a second topology analysis on the vessel topology information based on the vessel centerline according to a preset algorithm to generate a final segmentation result of the vessel.

According to the foregoing aspect and any possible implementation manner, an implementation manner is further provided, the second processing unit is specifically configured to:

calculating the parameter information of each vessel centerline corresponding to the vessel segment, and storing the vessel topology based on the vessel centerline in an undirected graph;

After the connected path is searched in the undirected graph according to the preset marker points, a second topology analysis is performed on the connected path through a preset algorithm based on the parameter information to generate a final segmentation result of the blood vessel.

According to the above aspect and any possible implementation, an implementation is further provided, the parameter information includes: the average radius of the blood vessel segment, the average tangent direction of the blood vessel segment, the radius at both ends of the blood vessel segment, and the tangent at both ends of the blood vessel segment One or more of the directions.

According to the foregoing aspect and any possible implementation manner, an implementation manner is further provided, wherein the first topology analysis and the second topology analysis include a vessel path analysis and a vessel bifurcation analysis.

According to the above aspect and any possible implementation, an implementation is further provided, the system further includes:

The storage unit is configured to respectively store the corresponding relationship between each blood vessel pixel point of the blood vessel and the blood vessel centerline.

Since each unit in this embodiment can execute the method shown in FIG. 1 , for parts not described in detail in this embodiment, reference may be made to relevant descriptions of FIGS. 1-3 .

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, server and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided by the present invention, it should be understood that the disclosed system, server and method can be implemented in other ways. For example, the server embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined Or it can be integrated into another system, or some features can be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of servers or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or in the form of hardware plus software functional units.

The above-mentioned integrated units implemented in the form of software functional units may be stored in a computer-readable storage medium. The above-mentioned software functional units are stored in a storage medium, and include several instructions to make a computer server (which can be a personal computer, a server, or a network server, etc.) or a processor (Processor) execute the methods described in various embodiments of the present invention. partial steps. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other various media that can store program codes. .

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the present invention. within the scope of protection.

Claims (10)

1. a kind of blood vessel separation method, it is characterised in that methods described includes：

The real-time tracing routing information of blood vessel original image is obtained by following the trail of；

First topological analysis is carried out to the real-time tracing routing information, the initial segmentation result of blood vessel is generated；

Vessel centerline is extracted according to the initial segmentation result；

According to preset algorithm, the second topological analysis is carried out to the blood vessel topology information based on the vessel centerline, blood vessel is generated Final segmentation result.

2. blood vessel separation method according to claim 1, it is characterised in that according to preset algorithm, to based on the blood vessel The blood vessel topology information of center line carries out the second topological analysis, generates the final segmentation result of blood vessel, specifically includes：

The parameter information of each vessel centerline correspondence vessel segment is calculated, and the blood vessel based on the vessel centerline is opened up Structure is flutterred to store with non-directed graph；

Searched according to preset mark point in the non-directed graph after communication path, preset algorithm pair is passed through based on the parameter information The communication path carries out the second topological analysis, generates the final segmentation result of blood vessel.

3. blood vessel separation method according to claim 2, it is characterised in that the parameter information includes：Vessel segment it is flat Equal radius, the average tangential direction of vessel segment, the one or more of vessel segment two ends radius and vessel segment two ends tangential direction.

4. blood vessel separation method according to claim 2, it is characterised in that first topological analysis and described second is opened up Flutterring analysis includes blood flow paths analysis and vascular bifurcation analysis.

5. blood vessel separation method according to claim 1, it is characterised in that methods described also includes：

Each blood vessel pixel of the blood vessel and the corresponding relation of vessel centerline are stored respectively.

6. blood vessel separation method according to claim 1, it is characterised in that the tracking uses ridgeline tracking, minimum One or more in cost path method, vessel cross-sections model tracking method and tubular structure model tracking method.

7. a kind of blood vessel piece-rate system, it is characterised in that the system includes：

Acquiring unit, the real-time tracing routing information for obtaining blood vessel original image by following the trail of；

First processing units, for carrying out the first topological analysis, initial point of generation blood vessel to the real-time tracing routing information Cut result；

Extraction unit, for extracting vessel centerline according to the initial segmentation result；

Second processing unit, for according to preset algorithm, second to be carried out to the blood vessel topology information based on the vessel centerline Topological analysis, generates the final segmentation result of blood vessel.

8. blood vessel piece-rate system according to claim 7, it is characterised in that the second processing unit, specifically for：

The parameter information of each vessel centerline correspondence vessel segment is calculated, and the blood vessel based on the vessel centerline is opened up Structure is flutterred to store with non-directed graph；

Searched according to preset mark point in the non-directed graph after communication path, preset algorithm pair is passed through based on the parameter information The communication path carries out the second topological analysis, generates the final segmentation result of blood vessel.

9. blood vessel piece-rate system according to claim 8, it is characterised in that first topological analysis and described second is opened up Flutterring analysis includes blood flow paths analysis and vascular bifurcation analysis.

10. blood vessel piece-rate system according to claim 7, it is characterised in that the system also includes：

Memory cell, for storing each blood vessel pixel of the blood vessel and the corresponding relation of vessel centerline respectively.