CN103247073A - Three-dimensional brain blood vessel model construction method based on tree structure - Google Patents

Abstract

The invention discloses a three-dimensional brain blood vessel model construction method based on a tree structure. The method includes the following steps: firstly, acquiring a three-dimensional brain blood vessel body data field from CT or MRA equipment; separating the brain blood vessel from background noise through a partitioning algorithm; secondly, calculating the skeleton line of the brain blood vessels, and building a tree-shaped brain blood vessel topological structure as per the skeleton line structure; next, calculating out the radius of the brain blood vessel at each control point through the elastic ball algorithm as per the skeleton line; and finally, three-dimensionally displaying the built brain blood vessels adopting the tree-shaped structure. The construction method provided by the invention accords with information of the space topological structure for brain blood vessels, and has the advantages of high vessel display precision and small result error; and the vessel lesion area can be detected, and the mapping in a multi-scale manner with respect to display windows of different sizes can be realized.

Description

Three-dimensional brain vessel model building method based on tree structure

Technical field

The invention belongs to medical domain, be specifically related to a kind of three-dimensional vascular pattern building method based on tree structure.

Background technology

Existing blood vessel Modeling Technology roughly can be divided into two classes: the method for unmounted model (model-free) and based on the method for model (model-based).

The most typical modal curve reestablishing method is MC(Marching Cubes in the method for unmounted model) algorithm.Thereby this method is divided into the purpose that two parts reach reconstruction by choosing an appropriate threshold and calculating a contour surface with the way of linear interpolation with the space.Yet linear interpolation use and that the space is divided into two-part way by threshold value is also too simple, therefore the effect of rebuilding is unsatisfactory.Need after rebuilding by smoothly eliminating the sawtooth effect of curved surface, simple Laplce's smoothing method can destroy tiny branch.Taubin has proposed the method for low-pass filtering, and Vollmer is, and Laplce smoothly improves, and has all obtained effect preferably.Constraint elastic surface net (CESN) has been obtained desirable effect by the accuracy that constrained in the unit under them preferably balance of the summit with level and smooth initial surface and slickness.

Method hypothesis blood vessel based on model is the structure of tubulose, the xsect that utilizes various geometric configuratioies and building method to approach blood vessel reaches the purpose of reconstruct blood vessel, commonly used is to adopt tubulose or spherical expression vascular pattern, this method not only defined in the three-dimensional entity model have a few, and explication its center line (skeleton), very being conducive to model is controlled in real time, is out of shape, is developed, is the building method that an extremely is suitable for expressing the such tubular articles of structure blood vessel.

Summary of the invention

Do not consider the blood vessel topological structure, in processing procedure, destroy the defective of tiny branch at existing method, the present invention proposes a kind of construction method of the three-dimensional brain vessel model based on tree structure, this model adopts tree structure to meet cerebrovascular topological property, adopt chou to close the center line mode and represent single vessel, this describing method can detect the cerebrovascular disease zone, and can adopt multiple dimensioned mode to draw at the display window of different sizes.

For achieving the above object, the present invention has adopted following technical scheme:

Three-dimensional vascular pattern building method based on tree structure may further comprise the steps:

(1) obtains three-dimensional cerebrovascular volume data field from CT or MRA equipment;

(2) adopt partitioning algorithm to separate the cerebrovascular and ground unrest;

(3) calculate cerebrovascular skeleton line;

(4) make up tree-shaped cerebrovascular topological structure according to skeleton line;

(5) adopt the elastic ball algorithm to calculate each node cerebrovascular radius according to skeleton line;

(6) the tree structure cerebrovascular that makes up is carried out 3-D display.

Preferably, described step (2) comprising:

(2.1) adopt gaussian filtering that smoothing processing is carried out in three-dimensional cerebrovascular volume data field, projection obtains the MIP image through MIP, adopts two-dimentional OTSU algorithm to obtain three-dimensional blood vessel seed point by the MIP image;

(2.2) define the region growing rule that overall situation and partial situation's information combines, by region growing algorithm blood vessel is carried out coarse segmentation, obtain region growing blood vessel profile;

(2.3) adopt the Catt diffusion model that anisotropic filtering is carried out in three-dimensional cerebrovascular volume data field, adopt local auto-adaptive C-V model, preliminary segmentation result is carried out secondary splitting as the initial profile line of self-adaptation movable contour model.

Preferably, be to adopt the method for Hessian matrix to calculate cerebrovascular tendency in the described step (3).

Preferably, described step (5) is the structure by elastic ball motoricity equation, seeks the elastic ball center line.

Preferably, described step (2.3) realizes by following formula:

$E^{\mathrm{LCV}}(c_1,c_2,c_1^{\′},c_2^{\′},\mathrm{\φ})=\mathrm{\μ}{\∫}_{\mathrm{\Ω}}\mathrm{\δ}\left(\mathrm{\φ}(x,y)\right)|\▿\mathrm{\φ}(x,y)|\mathrm{dxdy}+V{\∫}_{\mathrm{\Ω}}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+$

${\mathrm{\λ}}_1{\∫}_{\mathrm{\Ω}}|u_0(x,y)-c_1{|}^{2}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+{\mathrm{\λ}}_2{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_2|}^2(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}+$

${\mathrm{\λ}}_1^{\′}{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_1^{\′}|}^{2}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+{\mathrm{\λ}}_2^{\′}{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_2^{\′}|}^2(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}---\left(1\right)$

In formula (1), the formula (2),

$c_1=\frac{{\∫}_{{\mathrm{\Ω}}_1}{u}_0(x,y)H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_1}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}$ $c_2=\frac{{\∫}_{{\mathrm{\Ω}}_2}{u}_0(x,y)(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_2}(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}$

$c_1^{\′}=\frac{{\∫}_{{\mathrm{\Ω}}_1^{\′}}{u}_0(x,y)H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_1^{\′}}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}$ $c_2^{\′}=\frac{{\∫}_{{\mathrm{\Ω}}_2^{\′}}{u}_0(x,y)(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_2^{\′}}(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}.$

Preferably, the process of specifically asking for of described step (3) described acquisition Cerebral Vascular Clinic's line and radius information is as follows:

The elastic force equation is set, determines elastic ball center trend:

$F(i,j,k)=\underset{m=i-r}{\overset{i+r}{\mathrm{\Σ}}}\underset{n=j-r}{\overset{j+r}{\mathrm{\Σ}}}\underset{p=k-r}{\overset{k+r}{\mathrm{\Σ}}}\mathrm{Seg}(m,n,k)---\left(3\right)$

In the formula (3):

In the formula (4), V (m, n, k) expression segmentation result, Position (i, j, k) expression (i, j, the k) vector position of position pixel, r represents the vessel radius of current calculating, its computation process is as follows:

Step1: for current centerline points position end points, r=1 is set;

Step2: calculating elastic power;

Step3: if elastic force is zero, r=r+1 is set, repeats Step2, otherwise carry out next step;

Step4: the direction vector Mobility Center point along power, recomputate elastic force;

Step5: if elastic force is zero and two-phase occurs and cut that search finishes, otherwise next step;

Step6: center line and the radius information of the new revision of record, return Step3.

Preferably, described cerebrovascular tree topology structure construction method is as follows:

Adopt the node mode to represent each branch vessel structure, data structure definition is as follows:

Struct?Node

{

Father's node,

{ centerline position, radius, disc normal vector },

Left side child nodes,

Right child nodes

}

The node-classification strategy is as follows on the skeleton:

End points: the position that skeleton line begins or stops, the neighborhood relationships of end points is: have only an abutment points;

General point: constitute the fundamental point of skeleton line, exist two to face contact around the general point;

Bifurcation: the take-off point of skeleton line, bifurcation are bipartite textures, have three abutment points around the bifurcation;

The tree topology structure construction:

On the basis of above-mentioned node-classification, adopt the side rate of depth-first to make up the tree-shaped topological structure of blood vessel, algorithm is as follows:

Step1: be starting point with current observation node, directly set up a new tree construction, and deposit in as root node current;

Step2: enter next node to be seen;

Step3: current observation point is general point, joins in the node structure of handling, and returns Step2; Otherwise turn to Step4;

Step4: if observation point is bifurcation, set up two new nodes respectively as left child nodes and the right child nodes of this node, at left child nodes and the right child nodes, recurrence enters Step2 respectively;

Step5: after handling all nodes, the topological structure of vascular tree makes up and finishes.

Three-dimensional brain vessel model building method based on tree structure of the present invention not only can effectively be cut apart the thick branch of the cerebrovascular, and can also accurately extract cerebrovascular small structure.The three-dimensional brain vessel model that the present invention makes up adopts tree structure to meet cerebrovascular topological property, adopt chou to close the center line mode and represent single vessel, this describing method can detect the cerebrovascular disease zone, and can adopt multiple dimensioned mode to draw at the display window of different sizes.

Description of drawings

Fig. 1 is the process flow diagram of the three-dimensional brain vessel model building method based on tree structure of the present invention;

Fig. 2 separates the process flow diagram of the cerebrovascular and ground unrest for partitioning algorithm;

The result of the cerebrovascular skeleton line that Fig. 3 extracts;

Fig. 4 elastic ball algorithm obtains and puts radius corresponding on the center line;

The structural design of each node data in the tree-shaped cerebrovascular of Fig. 5;

The three-dimensional brain vessel model structure of Fig. 6 is the result show.

Embodiment

Below in conjunction with accompanying drawing the preferred embodiments of the present invention are described in detail, but the present invention is not restricted to these embodiment.The present invention is contained any in substituting of making of marrow of the present invention and scope, modification, equivalent method and scheme.Understand for the public is had completely the present invention, in the following preferred embodiment of the present invention, describe concrete details in detail, and do not have the description of these details also can understand the present invention fully for a person skilled in the art.

As shown in Figure 1, building method provided by the invention comprises following step:

Step S601: obtain the raw data of medical image, obtain three-dimensional cerebrovascular volume data field from CT or MRA equipment;

Step S602: adopt partitioning algorithm with blood vessel data and background separation, with blood vessel data and background separation.The specific implementation method of step S602 at first adopts gaussian filtering that smoothing processing is carried out in three-dimensional cerebrovascular volume data field as shown in Figure 2, and projection obtains the MIP image through MIP, adopts two-dimentional OTSU algorithm to obtain three-dimensional blood vessel seed point by the MIP image; Then, the region growing rule that adopts self-defined overall situation and partial situation information to combine is carried out coarse segmentation by region growing algorithm to blood vessel, obtains region growing blood vessel profile; At last, adopt the Catt diffusion model that anisotropic filtering is carried out in three-dimensional cerebrovascular volume data field, adopt local auto-adaptive C-V model, preliminary segmentation result is carried out secondary splitting as the initial profile line of self-adaptation movable contour model, obtain cerebrovascular skeleton line as shown in Figure 3.Described local auto-adaptive C-V model parameter arranges suc as formula shown in (1), the formula (2),

$E^{\mathrm{LCV}}(c_1,c_2,c_1^{\′},c_2^{\′},\mathrm{\φ})=\mathrm{\μ}{\∫}_{\mathrm{\Ω}}\mathrm{\δ}\left(\mathrm{\φ}(x,y)\right)|\▿\mathrm{\φ}(x,y)|\mathrm{dxdy}+V{\∫}_{\mathrm{\Ω}}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+$

${\mathrm{\λ}}_1{\∫}_{\mathrm{\Ω}}|u_0(x,y)-c_1{|}^{2}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+{\mathrm{\λ}}_2{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_2|}^2(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}+$

${\mathrm{\λ}}_1^{\′}{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_1^{\′}|}^{2}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+{\mathrm{\λ}}_2^{\′}{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_2^{\′}|}^2(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}---\left(1\right)$

Wherein,

$c_1=\frac{{\∫}_{{\mathrm{\Ω}}_1}{u}_0(x,y)H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_1}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}$ $c_2=\frac{{\∫}_{{\mathrm{\Ω}}_2}{u}_0(x,y)(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_2}(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}$

$c_1^{\′}=\frac{{\∫}_{{\mathrm{\Ω}}_1^{\′}}{u}_0(x,y)H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_1^{\′}}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}$ $c_2^{\′}=\frac{{\∫}_{{\mathrm{\Ω}}_2^{\′}}{u}_0(x,y)(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_2^{\′}}(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}$

Contour curve C is divided into interior zone Ω with image _InWith perimeter Ω _OutThe location of pixels of observing with current differentiation is round dot, and radius is that the ball of r is with Ω _InBe divided into Ω ₁, Ω ' ₁, with Ω _OutBe divided into Ω ₂, Ω ' ₂C ₁, C ₁', C ₂, C' ₂Represent regional Ω respectively ₁, Ω ' ₁, Ω ₂, Ω ' ₂Average.U ₀(x, y) expression point (x, y) average.

The variant function definition of H (φ):

$\mathrm{H\ϵ}=\frac{1}{2}(1+\frac{2}{\mathrm{\π}}\arctan \left(\frac{\mathrm{\φ}}{\mathrm{\ϵ}}\right))---(3-16)$

Experimental result shows that algorithm of the present invention not only can effectively cut apart the thick branch of the cerebrovascular, and can also accurately extract cerebrovascular small structure.

Step S603: calculate cerebrovascular skeleton line, obtain Cerebral Vascular Clinic's line and radius information, obtain on the center line and put radius corresponding as shown in Figure 4, employed is the elastic ball algorithm, and it is as follows that it specifically asks for process:

Elastic ball elastic force equation is set, determines elastic ball center trend:

$F(i,j,k)=\underset{m=i-r}{\overset{i+r}{\mathrm{\Σ}}}\underset{n=j-r}{\overset{j+r}{\mathrm{\Σ}}}\underset{p=k-r}{\overset{k+r}{\mathrm{\Σ}}}\mathrm{Seg}(m,n,k)---\left(3\right)$

In the formula (3),

R represents the vessel radius of current calculating,

In the formula (4), V (m, n, k) expression segmentation result, Position (i, j, k) expression (i, j, k) vector position of position pixel; Its computation process is as follows:

Step1: for current centerline points position end points, r=1 is set;

Step2: calculating elastic power;

Step3: if elastic force is zero, r=r+1 is set, repeats Step2, otherwise carry out next step;

Step4: the direction vector Mobility Center point along power, recomputate elastic force;

Step5: if elastic force is zero and two-phase occurs and cut that search finishes, otherwise next step;

Step6: center line and the radius information of the new revision of record, return Step3.

Step S604: make up tree-shaped cerebrovascular topological structure according to skeleton line.Concrete, adopt the node mode to represent the structure of each branch vessel, as shown in Figure 5, its data structure definition is as follows:

Struct?Node

{

Father's node,

{ centerline position, radius, disc normal vector },

Left side child nodes,

Right child nodes

}

The node of blood vessel is divided three classes: end points, general point and bifurcation are defined as follows:

End points: be positioned at the position that skeleton line begins or stops, the neighborhood relationships of end points is: have only an abutment points.

General point: constitute the fundamental point of skeleton line, exist two to face contact around the general point.

Bifurcation: the take-off point of skeleton line, because seldom there are many bifurcation structures in the cerebrovascular, so generally speaking, bifurcation is bipartite texture, bifurcation has three abutment points.

On the basis according to above-mentioned node-classification, adopt the side rate of depth-first to make up the tree-shaped topological structure of blood vessel, algorithm is as follows:

Step1: be starting point with current observation node, directly set up a new tree construction, and deposit in as root node current;

Step2: enter next node to be seen;

Step3: current observation point is general point, joins in the node structure of handling, and returns Step2; Otherwise turn to Step4;

Step4: if observation point is bifurcation, set up two new nodes respectively as left child nodes and the right child nodes of this node, at left child nodes and the right child nodes, recurrence enters Step2 respectively;

Step5: after handling all nodes, the topological structure of vascular tree makes up and finishes.

Step S605: according to skeleton line and adopt the elastic ball algorithm to calculate each node cerebrovascular radius, with reference to the detailed description of concrete steps (5).

Step S606: in computing machine that the vascular pattern that makes up is visual, show the result as shown in Figure 6.

The preferred embodiment of the present invention just is used for setting forth the present invention, and does not have all embodiments of detailed descriptionthe, and the present invention does not limit this invention and only is described embodiment.Obviously, according to the content of this instructions, can make many modifications and variations.These embodiment are chosen and specifically described to this instructions, is in order to explain principle of the present invention and practical application better, thereby the technical field technician can utilize the present invention well under making.The present invention only is subjected to the restriction of claims and four corner and equivalent.

Claims (7)

1. based on the three-dimensional brain vessel model building method of tree structure, it is characterized in that, may further comprise the steps:

(1) obtains three-dimensional cerebrovascular volume data field from CT or MRA equipment;

(2) adopt partitioning algorithm to separate the cerebrovascular and ground unrest;

(3) calculate cerebrovascular skeleton line;

(4) make up tree-shaped cerebrovascular topological structure according to skeleton line;

(5) adopt the elastic ball algorithm to calculate each node cerebrovascular radius according to skeleton line;

(6) the tree structure cerebrovascular that makes up is carried out 3-D display.

2. building method as claimed in claim 1 is characterized in that, described step (2) comprising:

(2.1) adopt gaussian filtering that smoothing processing is carried out in three-dimensional cerebrovascular volume data field, projection obtains the MIP image through MIP, adopts two-dimentional OTSU algorithm to obtain three-dimensional blood vessel seed point by the MIP image;

(2.2) define the region growing rule that overall situation and partial situation's information combines, by region growing algorithm blood vessel is carried out coarse segmentation, obtain region growing blood vessel profile;

(2.3) adopt the Catt diffusion model that anisotropic filtering is carried out in three-dimensional cerebrovascular volume data field, adopt local auto-adaptive C-V model, preliminary segmentation result is carried out secondary splitting as the initial profile line of self-adaptation movable contour model.

3. building method as claimed in claim 1 is characterized in that, in the described step (3) is to adopt the method for Hessian matrix to calculate cerebrovascular tendency.

4. building method as claimed in claim 1 is characterized in that, described step (5) is the structure by elastic ball motoricity equation, seeks the elastic ball center line.

5. building method as claimed in claim 2 is characterized in that, described step (2.3) realizes by following formula:

$E^{\mathrm{LCV}}(c_1,c_2,c_1^{\′},c_2^{\′},\mathrm{\φ})=\mathrm{\μ}{\∫}_{\mathrm{\Ω}}\mathrm{\δ}\left(\mathrm{\φ}(x,y)\right)|\▿\mathrm{\φ}(x,y)|\mathrm{dxdy}+V{\∫}_{\mathrm{\Ω}}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+$

${\mathrm{\λ}}_1{\∫}_{\mathrm{\Ω}}|u_0(x,y)-c_1{|}^{2}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+{\mathrm{\λ}}_2{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_2|}^2(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}+$

${\mathrm{\λ}}_1^{\′}{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_1^{\′}|}^{2}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}+{\mathrm{\λ}}_2^{\′}{\∫}_{\mathrm{\Ω}}{|u_0(x,y)-c_2^{\′}|}^2(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}---\left(1\right)$

In formula (1), the formula (2):

$c_1=\frac{{\∫}_{{\mathrm{\Ω}}_1}{u}_0(x,y)H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_1}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}$ $c_2=\frac{{\∫}_{{\mathrm{\Ω}}_2}{u}_0(x,y)(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_2}(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}$

$c_1^{\′}=\frac{{\∫}_{{\mathrm{\Ω}}_1^{\′}}{u}_0(x,y)H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_1^{\′}}H\left(\mathrm{\φ}(x,y)\right)\mathrm{dxdy}}$ $c_2^{\′}=\frac{{\∫}_{{\mathrm{\Ω}}_2^{\′}}{u}_0(x,y)(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}{{\∫}_{{\mathrm{\Ω}}_2^{\′}}(1-H\left(\mathrm{\φ}(x,y)\right))\mathrm{dxdy}}.$

6. building method as claimed in claim 3 is characterized in that, the process of specifically asking for of described step (3) described acquisition Cerebral Vascular Clinic's line and radius information is as follows:

The elastic force vector equation is set, determines elastic ball center trend:

$F(i,j,k)=\underset{m=i-r}{\overset{i+r}{\mathrm{\Σ}}}\underset{n=j-r}{\overset{j+r}{\mathrm{\Σ}}}\underset{p=k-r}{\overset{k+r}{\mathrm{\Σ}}}\mathrm{Seg}(m,n,k)$

In the formula (3):

R represents the vessel radius of current calculating,

In the formula (4), V (m, n, k) expression segmentation result, Position (i, j, k) expression (i, j, the k) vector position of position pixel, its computation process is as follows:

Step1: for current centerline points position end points, r=1 is set;

Step2: calculating elastic power;

Step3: if elastic force is zero, r=r+1 is set, repeats Step2, otherwise carry out next step;

Step4: the direction vector Mobility Center point along elastic force, recomputate elastic force;

Step5: if elastic force is zero and two-phase occurs and cut that search finishes, otherwise next step;

Step6: center line and the radius information of the new revision of record, return Step3.

7. building method as claimed in claim 2 is characterized in that, the construction method of described tree-shaped cerebrovascular topological structure is as follows:

Adopt the node mode to represent each branch vessel structure, data structure organization is as follows:

Struct?Node

{

Father's node,

{ centerline position, radius, disc normal vector },

Left side child nodes,

Right child nodes

}

Node-classification is as follows on the skeleton:

End points: the position that skeleton line begins or stops, the neighborhood relationships of end points is: have only an abutment points;

General point: constitute the fundamental point of skeleton line, exist two to face contact around the general point;

Bifurcation: the take-off point of skeleton line, bifurcation are bipartite textures, have three abutment points around the bifurcation;

The tree topology structure construction:

On the basis of above-mentioned node-classification, adopt the side rate of depth-first to make up the tree-shaped topological structure of blood vessel, algorithm is as follows:

Step1: be starting point with current observation node, directly set up a new tree construction, and deposit in as root node current;

Step2: enter next node to be seen;

Step3: current observation point is general point, joins in the node structure of handling, and returns Step2; Otherwise turn to Step4;

Step4: if observation point is bifurcation, set up two new nodes respectively as left child nodes and the right child nodes of this node, at left child nodes and the right child nodes, recurrence enters Step2 respectively;

Step5: after handling all nodes, the topological structure of vascular tree makes up and finishes.

Images