CN114677436A - Automatic cerebrovascular positioning method in brain CTA image based on model registration - Google Patents
Automatic cerebrovascular positioning method in brain CTA image based on model registration Download PDFInfo
- Publication number
- CN114677436A CN114677436A CN202210245464.3A CN202210245464A CN114677436A CN 114677436 A CN114677436 A CN 114677436A CN 202210245464 A CN202210245464 A CN 202210245464A CN 114677436 A CN114677436 A CN 114677436A
- Authority
- CN
- China
- Prior art keywords
- image
- model
- cerebrovascular
- central line
- center line
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010969 brain CTA Methods 0.000 title claims abstract description 73
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000002490 cerebral effect Effects 0.000 claims abstract description 73
- 238000013507 mapping Methods 0.000 claims abstract description 40
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 33
- 210000004556 brain Anatomy 0.000 claims abstract 3
- 230000006870 function Effects 0.000 claims description 34
- 230000009466 transformation Effects 0.000 claims description 16
- 230000004807 localization Effects 0.000 claims description 13
- 238000012935 Averaging Methods 0.000 claims description 7
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 6
- 238000012512 characterization method Methods 0.000 claims description 2
- 239000013589 supplement Substances 0.000 claims description 2
- 238000002372 labelling Methods 0.000 claims 1
- 230000003993 interaction Effects 0.000 description 7
- 238000003745 diagnosis Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 3
- 238000010968 computed tomography angiography Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 238000002059 diagnostic imaging Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 210000003625 skull Anatomy 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/68—Analysis of geometric attributes of symmetry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Abstract
The application relates to a brain blood vessel automatic positioning method in a brain CTA image based on model registration. The method comprises the following steps: mapping points in each floating image into a coordinate space of the fixed image through an image registration algorithm, and performing the same mapping operation on the cerebrovascular central line corresponding to each floating image to obtain a transformed cerebrovascular central line; calculating the average value of the gray scale of each point in the brain CTA image registered to the fixed image coordinate space to obtain a brain CTA model; applying a center line average algorithm to the transformed cerebrovascular center line to obtain a model center line; obtaining a cerebral vessel region of interest through a brain CTA model and a model centerline; the inertial characteristics of the blood vessel are obtained from the model central line mapped to the image to be positioned, the central line is guided in the region of interest to automatically position the cerebral vessels in the image, the positioning result of the central line of the cerebral vessels is output, and the automatic positioning of the cerebral vessels is realized.
Description
Technical Field
The application relates to the technical field of medical image processing, in particular to a model registration-based method for automatically positioning cerebral vessels in a brain CTA (computed tomography angiography) image.
Background
In 1895, the German physicist Louqin found that the appearance of medical images has a great influence on the traditional medical diagnosis mode. Nowadays, with the emergence and the vigorous development of medical imaging technologies such as fundus photography, magnetic resonance, computed tomography and the like, diagnosis based on medical images has become one of important auxiliary means in clinical diagnosis. The medical image can clearly and accurately represent the tissue distribution and pathological changes in the human body, and the information about the patient provides a lot of help for doctors to clinically diagnose the patient. Medical images are widely used at present.
Although medical imaging provides great convenience for diagnosis, it has some disadvantages.
At present, most imaging doctors judge the information presented by medical images according to own medical knowledge and clinical experience, and the analysis and diagnosis are carried out in such a way, so that the doctor level is highly required. Moreover, the method has low processing efficiency, wastes time and labor, and brings more and more workload to medical workers due to the medical images with huge data quantity. In addition, the body structures of different patients are basically the same, but still have a little difference, even medical images of the same part of the same patient in different periods can bring certain difficulty to diagnosis due to the difference of factors such as position, illumination and the like, and the experience and level of medical workers are tested. These causes lead to a considerable increase in the working intensity of the medical workers, with possible misjudgments with serious consequences.
In recent years, the research methods of blood vessel localization are mainly divided into a conventional method and a deep learning method, but the conventional method is mainly used for realizing the blood vessel localization of a brain CTA (CT angiography) image. Traditional methods for cerebrovascular localization include active contour models, level sets, multi-scale spatial filtering, etc. However, the current blood vessel localization method is difficult to locate the blood vessels due to the complex structure of the brain CTA image, and the common localization algorithm needs multiple interactions.
Disclosure of Invention
Therefore, it is necessary to provide an automatic cerebrovascular localization method in a brain CTA image based on model registration for the above technical problems, so as to solve the problems that the brain CTA image has a complex structure and is difficult to localize the cerebrovascular, and a general localization algorithm needs multiple interactions.
A method for automatic cerebrovascular localization in brain CTA images based on model registration, the method comprising:
selecting one of a plurality of brain CTA images as a fixed image, using the rest as floating images, mapping points in each floating image into a coordinate space of the fixed image through an image registration algorithm, and performing the same mapping operation on a cerebral vessel central line corresponding to each floating image to obtain a transformed cerebral vessel central line;
Calculating the average value of the gray scale of each point in the brain CTA image registered to the fixed image coordinate space to obtain a brain CTA model;
applying a center line average algorithm to the transformed cerebrovascular center line to obtain a model center line;
mapping the brain CTA model into an image to be positioned through an image registration algorithm, performing the same mapping on the center line of the model, mapping the center line of the model into the image to be positioned, obtaining a cerebral blood vessel region of interest with a fixed radius around the center line of the model mapped into the image to be positioned, and storing the generated region of interest;
obtaining the inertia characteristics of the blood vessel from the model central line mapped to the image to be positioned, integrating the inertia characteristics into a characteristic function of a rapid marching algorithm, guiding the central line in the region of interest to automatically position the cerebral vessel in the image, and outputting the positioning result of the central line of the cerebral vessel.
In one embodiment, the method further comprises:
and displaying the positioning result of the cerebral vessel central line on a display device by adopting three-dimensional display brain CTA image and vessel central line software.
In one embodiment, the step of selecting one of the brain CTA images as a fixed image, using the other brain CTA images as floating images, mapping points in the floating images into a coordinate space of the fixed image through an image registration algorithm, and performing the same mapping operation on a cerebrovascular centerline corresponding to the floating images to obtain a transformed cerebrovascular centerline includes:
Selecting one of a plurality of brain CTA images as a fixed image, and taking the rest images as floating images;
roughly aligning the fixed image and the floating image, registering the fixed image and the floating image by using rigid transformation, further registering by using affine transformation, then carrying out rough B-spline free deformation registration, then carrying out fine B-spline free deformation registration, continuously iterating and mapping points in the floating image to a coordinate space of the fixed image, and carrying out the same mapping operation on a cerebrovascular centerline corresponding to the floating image to obtain a transformed cerebrovascular centerline.
In one embodiment, the step of obtaining the model centerline by applying a centerline averaging algorithm to the transformed cerebrovascular centerlines comprises:
selecting a line with the minimum distance from other central lines as a middle central line;
calculating a cutting plane of each point of the middle center line to obtain intersection points of the cutting plane and other center lines;
then, averaging all the intersection points along the middle center line to obtain an average center line;
extending the average centerline, selecting the longest transformed cerebrovascular centerline, aligning its excess distal end with the end point of the average centerline, and supplementing the extended portion with the average centerline to obtain the model centerline.
In one embodiment, the step of acquiring the inertial feature of the blood vessel from the model centerline mapped to the image to be positioned, fusing the inertial feature into a feature function of a fast marching algorithm, guiding the centerline in the region of interest to automatically position the cerebral vessel in the image, and outputting the positioning result of the cerebral vessel centerline includes:
adding the inertial characteristic mapped to the model center line in the image to be positioned into the characteristic P of the blood vessel center line, and introducing the inertial characteristic into a function equation;
and (4) between the starting point and the end point of the manual annotation, combining a characteristic function of a rapid marching algorithm, performing forward propagation in the whole interested region, and outputting a positioning result of the central line of the cerebral vessel.
In one embodiment, the step of performing forward propagation in the entire region of interest between the manually labeled start point and end point in combination with the feature function of the fast marching algorithm to output the positioning result of the cerebrovascular vessel centerline includes:
establishing an energy model E by combining a characteristic function of a fast marching algorithm;
and through an energy model E, the forward propagation is started from the starting point between the manually marked starting point and the manually marked end point until the U value of each point in the region of interest is obtained, the minimum path energy from the starting point to the end point is deduced, and the positioning result of the central line of the cerebral blood vessel is output.
In one embodiment, the energy model E is:
E(C)＝∫Ω(P(C(t))+w)dt
wherein, C (t) represents a curve on the three-dimensional cerebrovascular vessel image, P (C (t)) represents the characteristic of the current position on the curve, Ω represents the length of the curve, the definition domain is [0, L ], L is the length of the cerebrovascular vessel, E (C) is the energy function on the curve path C, and w is a constant.
In one embodiment, the characteristic function is:
wherein, i (x) represents the inertia characteristic of the cerebral vessels, h (x) represents the voxel characteristic of the cerebral vessels, v (x) represents the enhancement of the cerebral vessels vesselness, epsilon is a positive number and is used for preventing the denominator from being zero, alpha, beta and gamma are respectively the indexes of i (x), h (x), v (x) and P (x) are characteristic functions.
According to the automatic positioning method of the cerebral vessels in the brain CTA image based on the model registration, one of the brain CTA images is selected as a fixed image, the rest of the brain CTA images are selected as floating images, points in each floating image are mapped into a coordinate space of the fixed image through an image registration algorithm, and the same mapping operation is carried out on the central lines of the cerebral vessels corresponding to each floating image to obtain the central lines of the cerebral vessels after transformation; calculating the average value of the gray scale of each point in the brain CTA image registered to the fixed image coordinate space to obtain a brain CTA model; applying a center line average algorithm to the transformed cerebrovascular center line to obtain a model center line; mapping the brain CTA model into an image to be positioned through an image registration algorithm, performing the same mapping on the center line of the model, mapping the center line of the model into the image to be positioned, obtaining a cerebral vessel region of interest with a fixed radius around the center line of the model mapped into the image to be positioned, and storing the generated region of interest; the method comprises the steps of obtaining the inertia characteristics of blood vessels from a model central line mapped to an image to be positioned, integrating the inertia characteristics into a characteristic function of a rapid marching algorithm, guiding the central line in an interested region to automatically position the cerebral vessels in the image, outputting the positioning result of the central line of the cerebral vessels, automatically establishing the interested region, further automatically positioning the cerebral vessels, and solving the problems that a brain CTA image is complex in structure, difficult to position the cerebral vessels, and a common positioning algorithm needs multiple interactions.
Drawings
FIG. 1 is a flow diagram of a method for automatic cerebrovascular localization in brain CTA images based on model registration in one embodiment;
fig. 2 is a schematic three-section view of the gray scale difference before registration of two brain CTA images;
fig. 3 is a schematic three-section view of the gray scale difference after registration of two brain CTA images;
FIG. 4 is a point cloud image of a region of interest of a blood vessel;
FIG. 5 is a diagram showing the positioning result of the centerline of the cerebral blood vessel.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.
In one embodiment, as shown in fig. 1, there is provided a method for automatic positioning of cerebral vessels in brain CTA image based on model registration, which is illustrated by applying the method to a terminal, and includes the following steps:
step S220, selecting one of the brain CTA images as a fixed image, using the rest as floating images, mapping points in each floating image to a coordinate space of the fixed image through an image registration algorithm, and performing the same mapping operation on the cerebrovascular center line corresponding to each floating image to obtain a transformed cerebrovascular center line.
The method comprises the steps of marking the cerebrovascular central line in a floating image in advance, wherein each floating image corresponds to the cerebrovascular central line. In one scene, the points in the floating image a are mapped into the coordinate space of the fixed image, and all mapping operations in the floating image a mapping process also act on the corresponding cerebral vessel center line of the floating image a. The image registration algorithm may be an affine transformation or a B-spline transformation, or the like.
Specifically, the method comprises the following steps: the image registration can carry out space geometric transformation on two or more images with different time and different shooting angles, so that points corresponding to the same position in space correspond one to one, and the aim of information fusion can be achieved. The two images to be registered are called Fixed Image (Fixed Image) and Image to be registered (Moving Image), respectively, by PfAnd PmThe registration process is a process of continuously iteratively finding the optimal transformation that points on the image to be registered are mapped on the fixed image. The mapping relationship at three-dimensional space coordinates (x, y, z) is as follows:
Pf(x,y,z)＝Pm(T(x,y,z))
where T (x, y, z) represents the spatial transformation at three-dimensional spatial coordinates (x, y, z).
The invention first roughly aligns the two images, then registers them using a rigid transformation, and then further registers using an affine transformation. After affine registration, rough B-spline free deformation registration is firstly carried out, and then fine B-spline free deformation registration is carried out. And continuously iterating to obtain final transformation, and acting the transformation on the marked central line to obtain the transformed cerebrovascular central line.
Step S240, calculating an average value of the gray levels of each point in the brain CTA image registered to the fixed image coordinate space to obtain a brain CTA model.
The brain CTA images are located in the same model space at the moment through image registration, and the brain CTA model is obtained by averaging the gray values of the same positions of the images.
And step S260, applying a center line average algorithm to the transformed cerebrovascular center line to obtain a model center line.
Specifically, the method comprises the following steps: the vessel centerlines of different brain CTA images are also mapped to the same model space, and a line with the minimum distance from other centerlines is selected first, which is called as the middle centerline. And calculating the cutting plane of each point of the middle center line to obtain the intersection point of the cutting plane and other center lines. These intersections are then averaged along the median centerline to obtain an average of the centerlines. Finally, the mean value of the central line obtained in the last step needs to be extended, because sometimes the mean value of the central line is possibly not long enough to guide the extraction of the central line in the image to be processed. To solve this problem, we first pick the longest centerline and then align its beyond distal end with the end point of the average centerline we obtained, the extension will be the complement of the average centerline. The centerline belonging to the model is now obtained.
A priori model is formed through the brain CTA model and the model center line, and due to the fact that the priori model has certain generalization characteristics, unknown brain CTA images to be processed can be registered with the priori model, so that the priori information of the cerebral vessels, including the inertial characteristics and the interested region of the cerebral vessels, can be obtained. And then combining the information with a fast-marching algorithm to realize the automatic positioning of the cerebral vessels.
Step S280, the brain CTA model is mapped into an image to be positioned through an image registration algorithm, the center line of the model is mapped in the image to be positioned in the same way, the center line of the model is mapped into the image to be positioned, a cerebrovascular region of interest is obtained in a fixed radius around the center line of the model mapped in the image to be positioned, and the generated region of interest is stored.
The image to be positioned is a brain CTA image to be processed, which is not subjected to cerebral vessel positioning. The brain CTA model is registered to an image to be positioned, the center line of the model is subjected to the same mapping, the center line of the model is used as an axis, and a region of interest of a cerebral vessel is established around the center line of the model by a fixed radius from the image to be positioned. Finally, the automatic positioning of the cerebral vessels in the region of interest is realized by using an inertial feature guided fast marching algorithm. The search area of the center line positioning algorithm can be greatly reduced through the interested area, and the automatic positioning efficiency of the cerebral vessels is improved.
And S300, acquiring the inertia characteristics of the blood vessel from the model center line mapped to the image to be positioned, integrating the inertia characteristics into a characteristic function of a rapid advance algorithm, guiding the center line in the region of interest to automatically position the blood vessel in the image, and outputting the positioning result of the center line of the blood vessel.
After the region of interest of the image to be positioned is obtained, inertial characteristics provided by the center line of the model are added into the characteristics P of the center line of the blood vessel and the starting point and the end point of the manual marking, and are introduced into a function Equation (Eikonal Equation), forward propagation is carried out in the whole region of interest by using a rapid marching algorithm between the starting point and the end point of the manual marking, and automatic positioning of the cerebral vessels is realized.
Wherein the function equation is:
where x is the current position, F (x) is the velocity of propagation at x, T (x) is a function of time to reach the current position x,
represents the gradient of T (x).
The method for automatically positioning the cerebral vessels in the brain CTA images based on the model registration comprises the steps of selecting one of the brain CTA images as a fixed image, using the rest of the brain CTA images as floating images, mapping points in each floating image into a coordinate space of the fixed image through an image registration algorithm, and performing the same mapping operation on the cerebral vessel center lines corresponding to each floating image to obtain the transformed cerebral vessel center lines; calculating the average value of the gray scale of each point in the brain CTA image registered to the fixed image coordinate space to obtain a brain CTA model; applying a center line average algorithm to the transformed cerebrovascular center line to obtain a model center line; mapping the brain CTA model into an image to be positioned through an image registration algorithm, performing the same mapping on the center line of the model, mapping the center line of the model into the image to be positioned, obtaining a cerebrovascular region of interest with a fixed radius around the center line of the model mapped into the image to be positioned, and storing the generated region of interest; the method comprises the steps of obtaining the inertia characteristics of blood vessels from a model center line mapped to an image to be positioned, integrating the inertia characteristics into a characteristic function of a rapid marching algorithm, guiding the center line in an interested region to automatically position the cerebral vessels in the image, outputting the positioning result of the cerebral vessel center line, automatically establishing the interested region, further automatically positioning the cerebral vessels, and solving the problems that a brain CTA image is complex in structure, difficult to position the cerebral vessels and a common positioning algorithm needs multiple interactions.
In one embodiment, the method further comprises:
the brain CTA image and the blood vessel center line software are displayed in three dimensions, so that the positioning result of the blood vessel center line is displayed on the display equipment.
The method comprises the steps of compiling software capable of displaying a brain CTA image and a blood vessel center line in a three-dimensional mode in advance, and displaying a positioning result of an image to be positioned and the blood vessel center line by using the software after a positioning result of the blood vessel center line is obtained. Visual Studio, C + +, Qt and the like can be used for compiling software for three-dimensionally displaying the brain CTA image and the blood vessel center line, after the positioning result of the cerebral vessel center line of the image to be positioned is obtained, the positioning result of the image to be positioned and the cerebral vessel center line is used as the input of the three-dimensional display software, and the automatic positioning of the cerebral vessel can be realized.
In one embodiment, the step of selecting one of the brain CTA images as a fixed image and the others as floating images, mapping points in the floating images into a coordinate space of the fixed image through an image registration algorithm, and performing the same mapping operation on the corresponding cerebrovascular centerlines of the floating images to obtain transformed cerebrovascular centerlines includes: selecting one of the brain CTA images as a fixed image and the rest as floating images; roughly aligning the fixed image with the floating image, registering the fixed image and the floating image by using rigid transformation, further registering by using affine transformation, then carrying out rough B-spline free deformation registration, then carrying out fine B-spline free deformation registration, continuously iterating and mapping points in the floating image to a coordinate space of the fixed image, and carrying out the same mapping operation on a cerebrovascular centerline corresponding to the floating image to obtain a transformed cerebrovascular centerline.
In one embodiment, the step of applying a centerline averaging algorithm to the transformed cerebrovascular centerline to obtain a model centerline comprises: selecting a line with the minimum distance from other central lines as a middle central line; calculating a cutting plane of each point of the middle center line to obtain intersection points of the cutting plane and other center lines; then, averaging all the intersection points along the middle center line to obtain an average center line; the average centerline is extended, the longest transformed cerebrovascular centerline is selected, the distal end beyond it is then aligned with the end point of the average centerline, and the extended portion will supplement the average centerline to obtain the model centerline.
In one embodiment, the step of obtaining the inertial feature of the blood vessel from the model centerline mapped to the image to be located, integrating the inertial feature into the feature function of the fast marching algorithm, guiding the centerline in the region of interest to automatically locate the cerebral vessel in the image, and outputting the locating result of the cerebral vessel centerline includes: adding the inertial characteristic of the model central line mapped to the image to be positioned into the characteristic P of the blood vessel central line, and introducing the characteristic into a equation of a function; and (4) between the manually marked starting point and the manually marked end point, combining a characteristic function of a fast advancing algorithm, carrying out forward propagation in the whole interested region, and outputting a positioning result of the central line of the cerebral vessel.
In one embodiment, the step of outputting the positioning result of the cerebrovascular centerline by performing forward propagation in the whole region of interest between the manually labeled start point and end point in combination with the feature function of the fast marching algorithm comprises: establishing an energy model E by combining a characteristic function of a fast marching algorithm; and through an energy model E, the forward propagation is started from the starting point between the manually marked starting point and the manually marked end point until the U value of each point in the region of interest is obtained, the minimum path energy from the starting point to the end point is deduced, and the positioning result of the central line of the cerebral blood vessel is output.
The feature P of the blood vessel central line has a smaller value near the feature of the blood vessel, a constant w is introduced to keep the expression of an energy model E positive, the positioning problem of the central line is converted into a problem of searching a global minimum value by searching a path with the minimum P + w integral, and the energy model E is as follows:
E(C)＝∫Ω(P(C(t))+W)dt
wherein, C (t) represents a curve on the three-dimensional cerebrovascular vessel image, P (C (t)) represents the characteristic of the current position on the curve, Ω represents the length of the curve, the definition domain is [0, L ], L is the length of the cerebrovascular vessel, E (C) is the energy function on the curve path C, and w is a constant.
From the starting point, forward propagation is performed until a U value is obtained for each point in the region of interest, and the minimum path energy from the starting point to the ending point can be derived, which is defined as:
Wherein U is defined as the minimum value of the path integral between the starting point and any point,
represents the distance between the starting point and the ending pointThere is a set of paths, C (t) represents the curve on the three-dimensional cerebrovascular image, P (C (t)) represents the feature of the current position on the curve, and U (P) is the minimum value of the integral of all paths between the starting point and the ending point.
In one embodiment, the characterization function is:
wherein i (x) represents the inertial feature of cerebrovascular vessel, h (x) represents the voxel feature of cerebrovascular vessel, v (x) represents the enhancement of cerebrovascular vesselness, epsilon is positive number for preventing denominator from being zero, alpha, beta, gamma are the indexes of i (x), h (x), v (x), and P (x) is the feature function.
The inertial characteristic i (x) of the cerebral vessels is obtained through the center line of the model, and in the process of tracking the central line of the cerebral vessels, the positioning results of the central line of the model and the central line of the cerebral vessels are divided into a certain number of sections for processing. When the first segment of the central line of the cerebral blood vessel is positioned, the characteristic function without inertia characteristics is used for positioning. The positioning of the other segments is based on the positioning result of the segment on the central line of the cerebral blood vessel, the corresponding central line of the model is mapped to the positioning result of the previous segment, and the same mapping is acted on the central line of the model of the current segment to guide the next positioning process.
Where e is a natural constant, d (x) is the distance between a given position in the fast marching algorithm and the start of the current segment, μdDistance, σ, of the starting point of the current segment from the guiding position provided for the mapped brain CTA model and model centerlinedThe parameters are normalized for distance.
The cerebrovascular vessel voxel characteristic h (x) is defined as the characteristic of the gray value of the voxel at each position in the region of interest of the image to be positioned.
Wherein M (x) is the current value of the voxel gray scale at a given position in the image to be located, mugThe mean value of the voxel gray levels of the whole region of interest of the image to be positioned.
Compared with the prior art that interaction actions such as manual marking of an interested region, marking of a starting point and an end point and the like are needed, the automatic positioning method for the cerebral vessels in the brain CTA image based on model registration can realize automatic positioning of the cerebral vessels by only marking the starting point and the end point for one time of interaction, and solves the problems that the cerebral vessels in the brain CTA image are large in number and complex in structure, the cerebral vessels are difficult to position, and a common positioning algorithm needs multiple interactions. The brain CTA model and the model center line are established to obtain the cerebrovascular interested region in the brain CTA image to be processed, and the inertial characteristics provided by the brain CTA model and the model center line are fused into the characteristic function of the rapid marching algorithm, so that the automatic positioning of the cerebrovascular is realized.
In order to further embody the beneficial effects of the above method for automatically positioning cerebral vessels in brain CTA images based on model registration, reference may be made to fig. 2-5, where fig. 2 illustrates that two brain CTA images are in the same coordinate space, and the skull contour of the two brain CTA images is not yet registered, and it can be seen that the difference in skull position is large. As can be seen from fig. 3, after the two brain CTA images are registered in the same coordinate space, the contours are overlapped, so that a better registration effect is achieved. As can be seen from fig. 4, the region of interest is obtained by intercepting the brain CTA image, and cerebrovascular localization is performed in the region of interest, so that the search area of the centerline localization algorithm can be greatly reduced, and the efficiency of automatic cerebrovascular localization is improved. As can be seen from fig. 5, the automatic positioning of the cerebrovascular centerline was successfully achieved.
It should be understood that, although the steps in the flowchart of fig. 1 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least some of the steps in fig. 1 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least some of the sub-steps or stages of other steps.
All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.
Claims (8)
1. A method for automatic cerebrovascular localization in brain CTA images based on model registration, the method comprising:
selecting one of a plurality of brain CTA images as a fixed image, using the rest as floating images, mapping points in each floating image into a coordinate space of the fixed image through an image registration algorithm, and performing the same mapping operation on a cerebrovascular centerline corresponding to each floating image to obtain a transformed cerebrovascular centerline;
Calculating the average value of the gray scale of each point in the brain CTA image registered to the fixed image coordinate space to obtain a brain CTA model;
applying a center line average algorithm to the transformed cerebrovascular center line to obtain a model center line;
mapping the brain CTA model into an image to be positioned through an image registration algorithm, performing the same mapping on the center line of the model, mapping the center line of the model into the image to be positioned, obtaining a cerebral blood vessel region of interest with a fixed radius around the center line of the model mapped into the image to be positioned, and storing the generated region of interest;
obtaining the inertia characteristics of the blood vessel from the model central line mapped to the image to be positioned, integrating the inertia characteristics into a characteristic function of a rapid marching algorithm, guiding the central line in the region of interest to automatically position the cerebral vessel in the image, and outputting the positioning result of the central line of the cerebral vessel.
2. The method of claim 1, further comprising:
and displaying the positioning result of the cerebral vessel central line on a display device by adopting three-dimensional display brain CTA image and vessel central line software.
3. The method according to claim 1, wherein the step of selecting one of the brain CTA images as a fixed image and the rest as floating images, mapping points in the floating images into the coordinate space of the fixed image by an image registration algorithm, and performing the same mapping operation on the corresponding brain vessel center lines of the floating images to obtain transformed brain vessel center lines comprises:
selecting one of a plurality of brain CTA images as a fixed image, and taking the rest images as floating images;
roughly aligning the fixed image and the floating image, registering the fixed image and the floating image by using rigid transformation, further registering by using affine transformation, then carrying out rough B-spline free deformation registration, then carrying out fine B-spline free deformation registration, continuously iterating and mapping points in the floating image to a coordinate space of the fixed image, and carrying out the same mapping operation on a cerebrovascular centerline corresponding to the floating image to obtain a transformed cerebrovascular centerline.
4. The method of claim 1, wherein the step of applying a centerline averaging algorithm to the transformed cerebrovascular centerlines to obtain a model centerline comprises:
Selecting a line with the minimum distance from other center lines as a middle center line;
calculating a cutting plane of each point of the middle center line to obtain intersection points of the cutting plane and other center lines;
then, averaging all the intersection points along the middle center line to obtain an average center line;
and extending the average central line, selecting the longest transformed cerebrovascular central line, aligning the far end exceeding the cerebrovascular central line with the end point of the average central line, and using the extended part as the supplement of the average central line to obtain the model central line.
5. The method according to claim 1, wherein the step of obtaining the inertial feature of the blood vessel from the model centerline mapped to the image to be located, integrating the inertial feature into a feature function of a fast marching algorithm, guiding the centerline in the region of interest to automatically locate the cerebral blood vessel in the image, and outputting the locating result of the cerebral blood vessel centerline comprises:
adding the inertial characteristic of the model central line mapped to the image to be positioned into the characteristic P of the blood vessel central line, and introducing the characteristic into a equation of a function;
and (4) between the manually marked starting point and the manually marked end point, combining a characteristic function of a fast advancing algorithm, carrying out forward propagation in the whole interested region, and outputting a positioning result of the central line of the cerebral vessel.
6. The method of claim 5, wherein the step of outputting the positioning result of the centerline of the cerebral blood vessel by forward propagation in the whole region of interest between the start point and the end point of the manual labeling in combination with the feature function of the fast marching algorithm comprises:
establishing an energy model E by combining a characteristic function of a fast marching algorithm;
and through an energy model E, the forward propagation is started from the starting point between the manually marked starting point and the manually marked end point until the U value of each point in the region of interest is obtained, the minimum path energy from the starting point to the end point is deduced, and the positioning result of the central line of the cerebral blood vessel is output.
7. The method of claim 6, wherein the energy model E is:
E(C)＝∫Ω(P(C(t))+w)dt
wherein, C (t) represents a curve on the three-dimensional cerebrovascular vessel image, P (C (t)) represents the characteristic of the current position on the curve, Ω represents the length of the curve, the definition domain is [0, L ], L is the length of the cerebrovascular vessel, E (C) is the energy function on the curve path C, and w is a constant.
8. The method of claim 6, wherein the characterization function is:
wherein, i (x) represents the inertia characteristic of the cerebral vessels, h (x) represents the voxel characteristic of the cerebral vessels, v (x) represents the enhancement of the cerebral vessels vesselness, epsilon is a positive number and is used for preventing the denominator from being zero, alpha, beta and gamma are respectively the indexes of i (x), h (x), v (x) and P (x) are characteristic functions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210245464.3A CN114677436A (en) | 2022-03-14 | 2022-03-14 | Automatic cerebrovascular positioning method in brain CTA image based on model registration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210245464.3A CN114677436A (en) | 2022-03-14 | 2022-03-14 | Automatic cerebrovascular positioning method in brain CTA image based on model registration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114677436A true CN114677436A (en) | 2022-06-28 |
Family
ID=82074602
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210245464.3A Pending CN114677436A (en) | 2022-03-14 | 2022-03-14 | Automatic cerebrovascular positioning method in brain CTA image based on model registration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114677436A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116309751A (en) * | 2023-03-15 | 2023-06-23 | 北京医准智能科技有限公司 | Image processing method, device, electronic equipment and medium |
| CN116487074A (en) * | 2023-06-20 | 2023-07-25 | 四川省医学科学院·四川省人民医院 | 5G-based remote medical assistance method, device and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190099145A1 (en) * | 2017-09-29 | 2019-04-04 | Infinitt Healthcare Co., Ltd. | Computing system and method for identifying and visualizing cerebral thrombosis based on medical images |
| CN113066091A (en) * | 2021-03-29 | 2021-07-02 | 昆明同心医联科技有限公司 | Cerebral vessel segmentation method and device based on black vessel wall curved surface reconstruction and storage medium |
| CN113947205A (en) * | 2021-10-13 | 2022-01-18 | 上海市第六人民医院 | Neural network model training method, computer-readable storage medium and device |
-
2022
- 2022-03-14 CN CN202210245464.3A patent/CN114677436A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190099145A1 (en) * | 2017-09-29 | 2019-04-04 | Infinitt Healthcare Co., Ltd. | Computing system and method for identifying and visualizing cerebral thrombosis based on medical images |
| CN113066091A (en) * | 2021-03-29 | 2021-07-02 | 昆明同心医联科技有限公司 | Cerebral vessel segmentation method and device based on black vessel wall curved surface reconstruction and storage medium |
| CN113947205A (en) * | 2021-10-13 | 2022-01-18 | 上海市第六人民医院 | Neural network model training method, computer-readable storage medium and device |
Non-Patent Citations (1)
| Title |
|---|
| LIU LIU等: "Automatic extraction of coronary centerline based on model-mapped and inertia-guided minimum path from CTA images", MULTIMED TOOLS APPL, vol. 78, 18 July 2018 (2018-07-18), pages 8767 - 8782, XP036770234, DOI: 10.1007/s11042-018-6335-4 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116309751A (en) * | 2023-03-15 | 2023-06-23 | 北京医准智能科技有限公司 | Image processing method, device, electronic equipment and medium |
| CN116487074A (en) * | 2023-06-20 | 2023-07-25 | 四川省医学科学院·四川省人民医院 | 5G-based remote medical assistance method, device and system |
| CN116487074B (en) * | 2023-06-20 | 2023-08-18 | 四川省医学科学院·四川省人民医院 | 5G-based remote medical assistance method, device and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11455732B2 (en) | Knowledge-based automatic image segmentation | |
| US9275432B2 (en) | Method of, and apparatus for, registration of medical images | |
| JP5129480B2 (en) | System for performing three-dimensional reconstruction of tubular organ and method for operating blood vessel imaging device | |
| US9342885B2 (en) | Method of generating a multi-modality anatomical atlas | |
| US8675944B2 (en) | Method of registering image data | |
| CN109410188B (en) | System and method for segmenting medical images | |
| CN107067398B (en) | Completion method and device for missing blood vessels in three-dimensional medical model | |
| US9466117B2 (en) | Segmentation highlighter | |
| CN114677436A (en) | Automatic cerebrovascular positioning method in brain CTA image based on model registration | |
| JPWO2007023723A1 (en) | Image processing method, image processing program, and image processing apparatus | |
| JP5934071B2 (en) | Apparatus, method and program for searching for shortest path of tubular structure | |
| KR20190084380A (en) | Method for extracting vascular structure in 2d x-ray angiogram, computer readable medium and apparatus for performing the method | |
| US9727975B2 (en) | Knowledge-based automatic image segmentation | |
| CN111161241A (en) | Liver image identification method, electronic equipment and storage medium | |
| US8938107B2 (en) | System and method for automatic segmentation of organs on MR images using a combined organ and bone atlas | |
| US20120169735A1 (en) | Improvements to curved planar reformation | |
| US9600918B2 (en) | Delineation and/or correction of a smooth stiff line in connection with an independent background image | |
| KR102229367B1 (en) | Cerebrovascular image displaying apparatus and method for comparison and diagnosis | |
| Cheirsilp | 3D multimodal image analysis for lung-cancer assessment | |
| Saeed et al. | A Robust Deterministic-Based Automatic Vessel Centerline Extraction Algorithm in 3-D Binary Volumes | |
| JP2024504025A (en) | container shape | |
| WO2022157124A1 (en) | Segment shape determination | |
| Kuehnel et al. | New analysis tools for the comprehensive assessment of the coronary arteries and myocardial viability in CT data sets | |
| 김지혜 | Locally Adaptive 2D-3D Registration using Vascular Structure Model for Liver Catheterization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |