CN116664644B - Vessel center line registration method and device based on magnetic resonance image - Google Patents
Vessel center line registration method and device based on magnetic resonance image Download PDFInfo
- Publication number
- CN116664644B CN116664644B CN202310935214.7A CN202310935214A CN116664644B CN 116664644 B CN116664644 B CN 116664644B CN 202310935214 A CN202310935214 A CN 202310935214A CN 116664644 B CN116664644 B CN 116664644B
- Authority
- CN
- China
- Prior art keywords
- blood vessel
- mapping
- point
- nuclear magnetic
- vessel
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 154
- 238000013507 mapping Methods 0.000 claims abstract description 152
- 230000011218 segmentation Effects 0.000 claims abstract description 45
- 238000012795 verification Methods 0.000 claims abstract description 19
- 238000000605 extraction Methods 0.000 claims description 29
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000013519 translation Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000008447 perception Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 4
- 239000008280 blood Substances 0.000 description 9
- 210000004369 blood Anatomy 0.000 description 9
- 230000009466 transformation Effects 0.000 description 8
- 238000003384 imaging method Methods 0.000 description 7
- 230000010349 pulsation Effects 0.000 description 7
- 239000013598 vector Substances 0.000 description 6
- 230000002792 vascular Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000005452 bending Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- 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
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- 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
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- 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
-
- 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
- G06T2207/30172—Centreline of tubular or elongated structure
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The application provides a registration method and a registration device based on a blood vessel center line of a magnetic resonance image, which belong to the technical field of medical image processing, and the method comprises the following steps: acquiring N nuclear magnetic image sequences; carrying out global rigid registration on the N nuclear magnetic image sequences; performing vessel segmentation on the N nuclear magnetic image sequences subjected to global rigid registration and extracting vessel center lines; selecting X characteristic points on the first blood vessel central line, searching X mapping characteristic points on the second blood vessel central line based on the X characteristic points, and registering the second blood vessel central line with the first blood vessel central line based on the X characteristic points and the X mapping characteristic points; checking the X characteristic points, and determining that the vessel center lines of the current reference nuclear magnetic image and the nuclear magnetic image to be registered are successfully registered under the condition that the checking is passed; in case the verification is not passed, the previous step is performed again. The application improves the accuracy and the robustness of registration and reduces the influence of the pulse and the thickness of the blood vessel on the registration.
Description
Technical Field
The application belongs to the technical field of medical image processing, and particularly relates to a vessel center line registration method and device based on a magnetic resonance image.
Background
The magnetic resonance technology is a scanning technology commonly used in clinic, a plurality of contrast images (i.e. imaging sequences) are often required to be acquired at different times in one magnetic resonance examination, and the images with the plurality of contrast images are often affected by the motion of a patient in the process of analyzing the images with the plurality of contrast images, so that the matching degree between the images with the plurality of contrast images is required to be registered, and the accuracy of a subsequent processing analysis link is further improved. Global registration based on images often has the problem that the registration effect is not ideal enough, so that the registration needs to be performed according to specific features of the imaging target, which are thinned to the images.
Vascular magnetic resonance imaging has been rapidly developed in recent years, and has gradually evolved from single bright blood imaging to an imaging scheme integrating bright blood imaging with a series of black blood imaging. Therefore, vessel registration based on multi-contrast images plays an increasingly important role in the analysis of clinical vessel magnetic resonance images. Vessel registration refers to matching a vessel image in a medical image with a real vessel for accurate vessel segmentation, measurement and navigation in clinical applications.
The conventional technical solutions of the present registration include registration based on feature points and registration based on geometric transformation. But there are the following problems in the application of these schemes to current vascular magnetic resonance images:
1. In the current registration scheme, gray level similarity of images with different contrast is often used as an evaluation standard of registration accuracy, but the vessel magnetic resonance image is often a multi-contrast image mixed by black blood and bright blood, and the gray levels of the two images have larger difference. Therefore, the registration difficulty of the bright blood and the black blood vessels based on the gray level similarity is high, and the accuracy is to be improved.
2. The blood vessel will have pulsations during imaging and the same vessel thickness and local position on different contrast images may differ. This poses a great difficulty in the selection of feature points in the registration process, and how to select feature points that are less affected by pulsations is a great challenge in vessel registration.
Disclosure of Invention
Based on the technical problems, the application provides a blood vessel center line registration method and device based on a magnetic resonance image, which reduces the influence of the blood vessel pulsation influence on the blood vessel registration result and improves the registration accuracy.
In a first aspect, the present application provides a method for registering a vessel centerline based on a magnetic resonance image, including:
step S1: acquiring N nuclear magnetic image sequences, wherein the N nuclear magnetic image sequences comprise 1 reference nuclear magnetic image sequence and N-1 nuclear magnetic image sequences to be registered;
Step S2: carrying out global rigid registration on the N nuclear magnetic image sequences;
step S3: performing blood vessel segmentation on the N nuclear magnetic image sequences subjected to global rigid registration to obtain a first blood vessel segmentation result corresponding to the reference nuclear magnetic image sequence and N-1 second blood vessel segmentation results corresponding to the N-1 nuclear magnetic image sequences to be registered;
step S4: extracting a first vessel center line in a first vessel segmentation result;
step S5: extracting a second vessel center line in a second vessel segmentation result of the kth nuclear magnetic image sequence to be registered, wherein k is more than or equal to 2 and less than or equal to N-1;
step S6: selecting X characteristic points on the first blood vessel central line, searching X mapping characteristic points on the second blood vessel central line based on the X characteristic points, and registering the second blood vessel central line with the first blood vessel central line based on the X characteristic points and the X mapping characteristic points;
step S7: checking the X feature points, and under the condition that the checking is passed, determining that the vessel center lines of the current reference nuclear magnetic image sequence and the kth nuclear magnetic image sequence to be registered are successfully registered, wherein k=k+1, and executing the step S5 again until the vessel center lines of all N-1 nuclear magnetic image sequences to be registered are successfully registered, and finishing the vessel center line registration; if the verification is not passed, step S6 is performed again.
The selection principle of the characteristic points comprises the following steps: and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel bifurcation as a characteristic point, and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel curvature on the first blood vessel centerline as a characteristic point at which the blood vessel curvature is larger than a preset curvature threshold.
The vascular bifurcation refers to a position where the blood vessels begin to extend in two different directions, respectively.
The blood vessel curvature refers to the bending degree of each position on the blood vessel central line.
The selecting X feature points on the first vessel centerline, searching X mapping feature points on the second vessel centerline based on the X feature points, and registering the second vessel centerline with the first vessel centerline based on the X feature points and the X mapping feature points includes:
step S6.1: selecting an ith characteristic point on the first blood vessel central line based on a selection principle, wherein i is more than or equal to 1 and less than or equal to X, and i is an integer;
step S6.2: mapping the ith feature point to a second blood vessel central line to obtain an ith mapping point;
step S6.3: the same selection principle as the ith feature point is adopted, points on the central line of a second blood vessel are searched in circles taking the ith mapping point as the circle center and the number of the first designated pixel points as the radius, and the searched points are determined to be the ith mapping feature points;
Step S6.4: selecting the i+1 th feature point by adopting the step S6.1, and searching by adopting the step S6.2 and the step S6.3 to obtain the i+1 th mapping feature point;
step S6.5: translating or rotating the second blood vessel center line by adopting the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point until the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point meet a first preset standard, and stopping translating or rotating the current second blood vessel center line;
step S6.6: and selecting the next characteristic point by adopting the step S6.1, and executing the steps S6.2-S6.5 again until the spatial relationship similarity of the X characteristic points meets a second preset standard, finishing the registration of the first blood vessel center line and the current second blood vessel center line, and determining the current second blood vessel center line as a registration result.
The method adopts the same selection principle as the ith feature point, searches the point on the second vessel center line in the circle with the ith mapping point as the circle center and the number of the first appointed pixel points as the radius, and determines the searched point as the ith mapping feature point, comprising the following steps:
Searching a point on the central line of the second blood vessel in a circle with the i-th mapping point as a circle center and the number of the first designated pixel points as a radius by adopting the same selection principle as the i-th characteristic point;
if a point located on the second vessel centerline can be searched, the searched point is used as an ith mapping feature point;
if the points on the second blood vessel central line cannot be searched, increasing the number of the specified pixel points, continuing to adopt the same selection principle as the ith feature point, searching the points on the second blood vessel central line in a circle with the ith mapping point as the circle center and the increased number of the specified pixel points as the radius until the points on the second blood vessel central line are searched, and taking the points obtained by searching at the moment as the ith mapping feature point.
The spatial position relationship between the ith feature point and the (i+1) th feature point and the spatial position relationship between the ith mapping feature point and the (i+1) th mapping feature point conform to a first preset standard, and the method comprises the following steps: the Euclidean distance between the ith feature point and the (i+1) th feature point and between the ith mapping feature point and the (i+1) th mapping feature point is less than or equal to a first Euclidean distance threshold, and the cosine similarity between the ith feature point and the (i+1) th feature point and between the (i mapping feature point and the (i+1) th mapping feature point is less than or equal to a first cosine similarity threshold.
The value rule of X comprises: if the number of the mapping feature points required to be obtained on the central line of each second blood vessel in the nuclear magnetic image sequence to be registered is not less thanpIf the value rule of X is thatWherein->For the number of vessels contained in the kth sequence of nuclear magnetic images to be registered, the reference sequence of nuclear magnetic images and the reference sequence of nuclear magnetic imagesEach blood vessel in the nuclear magnetic image sequence to be registered can be extracted to obtain a blood vessel center line.
The verifying the X feature points includes:
reconstructing a cross section of a blood vessel where the current first blood vessel center line is located at the ith characteristic point;
in the cross section, taking the ith characteristic point as a circle center and taking the number of second designated pixel points as a radius to make a circle;
performing similarity calculation on gray values of the reference nuclear magnetic image sequence and the nuclear magnetic image sequence to be registered in the circle by adopting a perception hash algorithm to obtain hash values of the reference nuclear magnetic image sequence and hash values of the nuclear magnetic image sequence to be registered at positions corresponding to all pixels in the circle;
comparing the hash value of the reference nuclear magnetic image sequence at the corresponding positions of all pixels in the circle with the hash value of the nuclear magnetic image sequence to be registered; and if the number of the hash values of the corresponding positions is not equal to or less than the first similarity threshold, checking passing, otherwise, checking failing.
In a second aspect, the present application proposes a vessel centerline registration device based on magnetic resonance images, comprising:
the image acquisition module is used for acquiring N nuclear magnetic image sequences, wherein the N nuclear magnetic image sequences comprise 1 reference nuclear magnetic image sequence and N-1 nuclear magnetic image sequences to be registered;
the global registration module is used for carrying out global rigid registration on the N nuclear magnetic image sequences;
the vessel segmentation module is used for carrying out vessel segmentation on the N nuclear magnetic image sequences subjected to global rigid registration to obtain a first vessel segmentation result corresponding to the reference nuclear magnetic image sequence and a second vessel segmentation result corresponding to the N-1 nuclear magnetic image sequences to be registered;
the first central line extraction module is used for extracting a first blood vessel central line in the first blood vessel segmentation result;
the second central line extraction module is used for extracting a second blood vessel central line in a second blood vessel segmentation result of the kth nuclear magnetic image sequence to be registered, wherein k is more than or equal to 2 and less than or equal to N-1;
the local registration module is used for selecting X characteristic points on the first blood vessel central line, searching X mapping characteristic points on the second blood vessel central line based on the X characteristic points, and registering the second blood vessel central line with the first blood vessel central line based on the X characteristic points and the X mapping characteristic points;
The verification module is used for verifying the X feature points, under the condition that verification is passed, determining that the vessel center lines of the current reference nuclear magnetic image sequence and the kth nuclear magnetic image sequence to be registered are successfully registered, wherein k=k+1, and executing the second center line extraction module again until the vessel center lines of all N-1 nuclear magnetic image sequences to be registered are successfully registered, and finishing the vessel center line registration; in case the verification is not passed, the registration module is executed again.
In a third aspect, the present application proposes an electronic device comprising: the system comprises a memory and a processor, wherein the memory stores a computer program which is executed by the processor to execute the vessel center line registration method based on the magnetic resonance image.
In a fourth aspect, the present application proposes a computer readable storage medium storing executable instructions that when executed cause a processor to perform the described magnetic resonance image based vessel centerline registration method.
The beneficial effects are that:
the application provides a registering method and a registering device based on a blood vessel center line of a magnetic resonance image, which adopt a registering mode of firstly wholly and then locally, improve the matching degree of characteristic points and reduce the problem of inaccurate selection. The registration mode based on the blood vessel center line is adopted, so that the influence of the surrounding tissues of the blood vessel, the pulsation of the blood vessel and the thickness of the blood vessel on the registration is reduced. The accuracy and the robustness of registration are improved, and the method is applicable to more registration scenes.
Drawings
FIG. 1 is a flowchart of a registration method based on a vessel centerline of a magnetic resonance image according to an embodiment of the present application;
FIG. 2 is a partial registration flow chart of an embodiment of the present application;
fig. 3 is a schematic block diagram of a registration device based on a blood vessel centerline of a magnetic resonance image according to an embodiment of the present application.
Detailed Description
The disclosure is further described below with reference to the embodiments shown in the drawings.
Vessel registration refers to matching a vessel image in a medical image with a real vessel, and is usually performed based on registration of feature points or based on geometric transformation in the vessel image in the prior art. The common method for feature point-based registration is as follows: by extracting key points in the vessel image and using these key points for matching and alignment. The defects are as follows: 1. the accuracy of feature point selection is limited by the vessel image quality and noise level. If there is a lot of noise in the vessel image or the image quality is poor, the extracted feature points may be inaccurate, resulting in an unsatisfactory registration result. 2. The location and direction of the feature points may not be unique. Due to factors such as different blood vessel image acquisition devices, different shooting angles and the like, the positions and directions of the same feature point in different blood vessel images may be different, and errors can occur in matching. 3. The robustness of feature point matching is poor. For some non-linearly deformed or rotated vessel images, feature point matching may fail, resulting in registration failure. The geometric transformation-based registration method is typically: one vessel image is mapped onto the other vessel image by calculating a geometric transformation matrix between the two vessel images. The defects are as follows: 1. the computational complexity is high. For large-scale vessel image sequences, computing the geometric transformation matrix requires a significant amount of time and computational resources. 2. There may be errors in the registration results. Due to geometrical transformation uncertainties, registration results may be subject to certain errors, especially when there is complex deformation or rotation in the vessel image. 3. The adaptability is poor. The registration method based on geometric transformation can only register static blood vessel images, and has poor registration effect on dynamic images or images containing moving elements.
The application provides a blood vessel center line registration method and a device based on a magnetic resonance image, which adopt a registration mode of firstly carrying out global rigid registration on the blood vessel image and then carrying out local registration on the blood vessel center line, thereby improving the matching degree of the characteristic points and reducing the inaccurate selection. The registration mode based on the blood vessel center line is adopted, so that the influence of the surrounding tissues of the blood vessel, the pulsation of the blood vessel and the thickness of the blood vessel on the registration is reduced. The accuracy and the robustness of registration are improved, and the method is applicable to more registration scenes.
Example 1:
the present embodiment proposes a vessel centerline registration method based on magnetic resonance images, as shown in fig. 1, including:
step S1: acquiring N nuclear magnetic image sequences, wherein the N nuclear magnetic image sequences comprise 1 reference nuclear magnetic image sequence and N-1 nuclear magnetic image sequences to be registered;
in the prior art, the blood vessel registration is usually carried out directly on the basis of the bright blood or black blood images with gray level similarity, so that the difficulty is high, and the accuracy is to be improved. And the registration of the feature points selected based on the bright or dark blood images is affected by the vessel pulsations. In this embodiment, N nuclear magnetic image sequences are directly acquired, and gray-scale processing is not required, and the N nuclear magnetic image sequences may be static vessel nuclear magnetic images or dynamic vessel nuclear magnetic images, so this embodiment may be applicable to more registration scenes. Each nuclear magnetic image sequence (comprising 1 reference nuclear magnetic image sequence and N-1 nuclear magnetic image sequences to be registered) is provided with M nuclear magnetic images, N is a positive integer greater than 1, and M is a positive integer.
Step S2: carrying out global rigid registration on the N nuclear magnetic image sequences;
in this embodiment, global rigid registration is required to be performed on all the nuclear magnetic image sequences, so as to primarily reduce image deviation. The present embodiment adopts "B-spline image registration method" as global registration algorithm. The algorithm is an image registration method based on B spline interpolation, can register images under different coordinate systems, and has high precision and high robustness.
Step S3: performing blood vessel segmentation on the N nuclear magnetic image sequences subjected to global rigid registration to obtain a first blood vessel segmentation result corresponding to the reference nuclear magnetic image sequence and N-1 second blood vessel segmentation results corresponding to the N-1 nuclear magnetic image sequences to be registered;
in this embodiment, the vessel segmentation needs to be performed on the N nuclear magnetic image sequences after global rigid registration, respectively. The segmentation method adopts the traditional threshold segmentation or manual adjustment mode to ensure the accuracy of blood vessel segmentation.
Step S4: extracting a first vessel center line in a first vessel segmentation result;
step S5: extracting a second vessel center line in a second vessel segmentation result of the kth nuclear magnetic image sequence to be registered, wherein k is more than or equal to 2 and less than or equal to N-1, and k and N are positive integers;
In this embodiment, extraction of the vessel center line is completed by adopting a skeleton-based refinement algorithm, and the existing vessel center line extraction algorithm can be divided into two types, namely, based on skeleton refinement, a refined skeleton is obtained mainly based on morphological corrosion operation, and a center line is finally obtained; and secondly, based on distance transformation, the main principle is that a shortest path searching algorithm is utilized to obtain a shortest path in a distance field, and a correction algorithm is combined to extract a center line. In this embodiment, the extraction of the vessel centerline is selected based on a skeleton refinement algorithm, considering good compatibility with the whole morphological analysis system.
It should be noted that: a second vessel segmentation result can be segmented in one nuclear magnetic image sequence to be registered, and a second vessel center line can be extracted from one second vessel segmentation result, so that N-1 nuclear magnetic image sequences to be registered correspondingly obtain N-1 second vessel center lines.
Step S6: selecting X characteristic points on the first blood vessel central line, searching X mapping characteristic points on the second blood vessel central line based on the X characteristic points, and registering the second blood vessel central line with the first blood vessel central line based on the X characteristic points and the X mapping characteristic points;
The selection principle of the characteristic points comprises the following steps: and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel bifurcation as a characteristic point, and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel curvature on the first blood vessel centerline as a characteristic point at which the blood vessel curvature is larger than a preset curvature threshold.
The vascular bifurcation refers to a position where the blood vessels begin to extend in two different directions, respectively.
The blood vessel curvature refers to the bending degree of each position on the blood vessel central line. The mathematical angle interpretation, that is, the rotation rate of the tangential direction angle to the arc length for a certain point on the curve (the curve refers to the central line of the blood vessel in the embodiment), is defined by differentiation, and indicates the degree of deviation of the curve from the straight line. Mathematically, a numerical value indicating the degree of curve bending at a certain point. The larger the curvature, the greater the degree of curvature of the curve. The calculation formula is as follows: for a smooth function f (x), its tangential spatial derivative d2f/dx2 at point x can be expressed as: d2f/dx2= (dF/dx)h, where h is the tangential plane normal vector of the function at x and dF/dx is the second partial derivative of the function at that point.
In this embodiment, the feature points at the bifurcation of the blood vessel are preferentially selected, and then the feature points with curvature greater than 0.5 are selected, wherein 0.5 is a predetermined curvature threshold value, and the adjustment can be performed according to specific needs.
X feature points are selected on the first blood vessel center line, X mapping feature points are searched on the second blood vessel center line based on the X feature points, and the second blood vessel center line is registered with the first blood vessel center line based on the X feature points and the X mapping feature points, as shown in fig. 2, specifically comprising:
step S6.1: selecting an ith characteristic point on the first blood vessel central line based on a selection principle, wherein i is more than or equal to 1 and less than or equal to X, and i is an integer;
in this embodiment, the feature point selection sequence is: the human body coordinate system is used as a reference coordinate system, and the human body is searched according to the sequence from foot to head, front to back and left to right. The value rule of X comprises: if the number of the mapping feature points required to be obtained on the central line of each second blood vessel in the nuclear magnetic image sequence to be registered is not less thanpIf the number is X, the value of XThe rule is p×y, where the embodimentpAnd taking 3, wherein Y is the number of blood vessels contained in the kth nuclear magnetic image sequence to be registered, and extracting each blood vessel in the reference nuclear magnetic image sequence and each blood vessel in the nuclear magnetic image sequence to be registered to obtain a blood vessel center line. The number of blood vessels contained in each segmented nuclear magnetic image sequence to be registered may be the same or different, for example the number of blood vessels contained in the first nuclear magnetic image sequence to be registered is 3, p=3，X is p×y=9, as follows: the number of vessels contained in the first sequence of nuclear magnetic images to be registered is 5,p=3，in the specific implementation, a first feature point is selected, then a first mapping feature point corresponding to the first feature point is searched, then a second feature point is selected in step S6.1, then the second vessel centerline is translated or rotated based on the first feature point, the first mapping feature point, the second feature point and the second mapping feature point until the spatial relationship between the first feature point, the first mapping feature point, the second feature point and the second mapping feature point meets a first preset standard, the translation or rotation of the second vessel centerline is stopped, the third feature point is reselected in step S6.1, and a third mapping feature point is searched and obtained, and the second vessel centerline is translated or rotated based on the second feature point, the second mapping feature point and the third mapping feature point in turn until the spatial relationship similarity between the X feature points meets a second preset standard, and the current second vessel centerline is determined to be a registration result by analogy.
Step S6.2: mapping the ith feature point to a second blood vessel central line to obtain an ith mapping point;
it should be noted that the i-th mapping point obtained at this time is not the i-th mapping feature point, but is just a rough mapping point, and is not the mapping feature point, but the accurate mapping feature point can be determined after searching within a specified range.
Step S6.3: the same selection principle as the ith feature point is adopted, points on the central line of a second blood vessel are searched in circles taking the ith mapping point as the circle center and the number of the first designated pixel points as the radius, and the searched points are determined to be the ith mapping feature points;
in this embodiment, the same selection principle as the ith feature point is adopted, specifically, if the ith feature point adopts the principle that the curvature is greater than 0.5, then when searching the ith mapping feature point, the principle that the curvature is greater than 0.5 is also adopted, and a point on the central line of the second blood vessel is searched in a circle with the ith mapping point as the center and the number of the first designated pixel points as the radius; in this embodiment, the number of the first designated pixels is 15 pixels.
If a point located on the second vessel centerline can be searched, the searched point is used as an ith mapping feature point;
If the points on the second blood vessel central line cannot be searched, increasing the number of the designated pixel points, continuing to adopt the same selection principle (namely the principle that the curvature is larger than 0.5) as the ith feature point, searching the points on the second blood vessel central line in a circle taking the ith mapping point as the circle center and taking the number of the increased designated pixel points as the radius until the points on the second blood vessel central line are searched, and taking the points obtained by searching at the moment as the ith mapping feature points. In this embodiment, the number of the designated pixels is increased to 5 pixels, and in a specific implementation, 5 pixels may be enlarged at a time.
Step S6.4: selecting the i+1 th feature point by adopting the step S6.1, and searching by adopting the step S6.2 and the step S6.3 to obtain the i+1 th mapping feature point;
step S6.5: translating or rotating the second blood vessel center line by adopting the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point until the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point meet a first preset standard, and stopping translating or rotating the current second blood vessel center line;
Step S6.6: and selecting the next characteristic point by adopting the step S6.1, and executing the steps S6.2-S6.5 again until the spatial relationship similarity of the X characteristic points meets a second preset standard, finishing the registration of the first blood vessel center line and the current second blood vessel center line, and determining the current second blood vessel center line as a registration result.
The spatial position relationship between the ith feature point and the (i+1) th feature point and the spatial position relationship between the ith mapping feature point and the (i+1) th mapping feature point conform to a first preset standard, and the method comprises the following steps: the Euclidean distance between the ith feature point and the (i+1) th feature point and between the ith mapping feature point and the (i+1) th mapping feature point is less than or equal to a first Euclidean distance threshold, and the cosine similarity between the ith feature point and the (i+1) th feature point and between the (i mapping feature point and the (i+1) th mapping feature point is less than or equal to a first cosine similarity threshold.
The Euclidean distance d is calculated as follows:
for points (x 1 , y 1 ) Sum point (x) 2 , y 2 ) The euclidean distance of (2) is:
in the present embodiment, x 1 As the ith feature point, y 1 Is the (i+1) th feature point, x 2 For the ith mapping feature point, y 2 Feature points are mapped for the i+1 th.
Likewise, for a point (x 1 , y 1 , z 1 ) Sum point (x) 2 , y 2 , z 2 ) Their euclidean distance is:
in the present embodiment, x 1 As the ith feature point, y 1 Z as the (i+1) th feature point 1 As the (i+2) th feature point, x 2 For the ith mapping feature point, y 2 Mapping feature points for the (i+1) th, z 2 Feature points are mapped for the i+2 th.
The cosine similarity is calculated by the formula: assuming that two vectors a and b, both of which are n in length, are provided, their cosine similarity is:
in this embodiment, the vector a is a vector formed by pointing the i-th feature point to the i+1-th feature point, and the vector b is a vector formed by pointing the i-th mapping feature point to the i+1-th mapping feature point.
Step S7: checking the X feature points, and under the condition that the checking is passed, determining that the vessel center lines of the current reference nuclear magnetic image sequence and the kth nuclear magnetic image sequence to be registered are successfully registered, wherein k=k+1, returning to and executing step S5 (automatically increasing k by 1, namely extracting the second vessel center line in the second vessel segmentation result of the kth+1th nuclear magnetic image sequence to be registered) until the vessel center lines of all N-1 nuclear magnetic image sequences to be registered are successfully registered, and finishing the vessel center line registration; if the verification is not passed, the step S6 is returned and executed again, i.e. the X feature points are selected again.
The verifying the X feature points includes:
reconstructing a cross section of a blood vessel where the current first blood vessel center line is located at the ith characteristic point;
in the cross section, taking the ith characteristic point as a circle center and taking the number of second designated pixel points as a radius to make a circle; in this embodiment, the number of the second designated pixels is 30 pixels, and it can be understood that the number of the second designated pixels can be valued according to the specific precision requirement.
Performing similarity calculation on gray values of the reference nuclear magnetic image sequence and the nuclear magnetic image sequence to be registered in the circle by adopting a perception hash algorithm to obtain hash values of the reference nuclear magnetic image sequence and hash values of the nuclear magnetic image sequence to be registered at positions corresponding to all pixels in the circle;
comparing the hash value of the reference nuclear magnetic image sequence at the corresponding positions of all pixels in the circle with the hash value of the nuclear magnetic image sequence to be registered; and if the number of the hash values of the corresponding positions is not equal to or less than the first similarity threshold, checking passing, otherwise, checking failing.
It can be understood that the hash values of the corresponding positions are much the same, which indicates that the reference nuclear magnetic image sequence and the nuclear magnetic image sequence to be registered are similar, whereas the hash values of the corresponding positions are much less, which indicates that the reference nuclear magnetic image sequence and the nuclear magnetic image sequence to be registered are different, so in this embodiment, the first similarity threshold is taken as 5, that is, when the number of the hash values of the corresponding positions is equal to or less than 5, which indicates that the reference nuclear magnetic image sequence and the nuclear magnetic image sequence to be registered are similar, and the verification passes, whereas the verification does not pass.
The vessel center line registration method based on the magnetic resonance image, provided by the embodiment, adopts the concept of firstly integrating and then locally, carries out global rigid registration on all nuclear magnetic image sequences, then extracts the vessel center line from the nuclear magnetic image sequences subjected to global rigid registration, realizes registration of the quasi-nuclear magnetic image sequence and the nuclear magnetic image sequences to be registered by extracting the characteristic points on the vessel center line, finally, needs to verify all the characteristic points, and if verification does not pass, needs to extract the proper characteristic points on the vessel center line again. The method provided by the embodiment improves the registration accuracy and is applicable to more registration scenes. The registration mode of firstly wholly and then locally improves the matching degree of the characteristic points and reduces the inaccurate selection problem. The registration mode based on the blood vessel center line is adopted, so that the influence of the surrounding tissues of the blood vessel, the pulsation of the blood vessel and the thickness of the blood vessel on the registration is reduced. Registration accuracy and robustness are improved.
Example 2:
the present embodiment proposes a vessel centerline registration device based on magnetic resonance image, as shown in fig. 1, including: the system comprises an image acquisition module, a global registration module, a blood vessel segmentation module, a central line extraction module, a local registration module and a verification module;
The image acquisition module is connected with the global registration module, the global registration module is connected with the blood vessel segmentation module, the blood vessel segmentation module is connected with the first central line extraction module, the first central line extraction module is connected with the second central line extraction module, the second central line extraction module is connected with the local registration module, the local registration module is connected with the verification module, and the verification module is respectively connected with the local registration module and the second central line extraction module;
the image acquisition module is used for acquiring N nuclear magnetic image sequences, wherein the N nuclear magnetic image sequences comprise 1 reference nuclear magnetic image sequence and N-1 nuclear magnetic image sequences to be registered;
the global registration module is used for carrying out global rigid registration on the N nuclear magnetic image sequences;
the vessel segmentation module is used for carrying out vessel segmentation on the N nuclear magnetic image sequences subjected to global rigid registration to obtain a first vessel segmentation result corresponding to the reference nuclear magnetic image sequence and a second vessel segmentation result corresponding to the N-1 nuclear magnetic image sequences to be registered;
the first central line extraction module is used for extracting a first blood vessel central line in the first blood vessel segmentation result;
The second central line extraction module is used for extracting a second blood vessel central line in a second blood vessel segmentation result of the kth nuclear magnetic image sequence to be registered, wherein k is more than or equal to 2 and less than or equal to N-1;
the local registration module is used for selecting X characteristic points on the first blood vessel central line, searching X mapping characteristic points on the second blood vessel central line based on the X characteristic points, and registering the second blood vessel central line with the first blood vessel central line based on the X characteristic points and the X mapping characteristic points;
the verification module is used for verifying the X feature points, under the condition that verification is passed, determining that the vessel center lines of the current reference nuclear magnetic image sequence and the kth nuclear magnetic image sequence to be registered are successfully registered, wherein k=k+1, and executing the second center line extraction module again until the vessel center lines of all N-1 nuclear magnetic image sequences to be registered are successfully registered, and finishing the vessel center line registration; in case the verification is not passed, the registration module is executed again.
The local registration module comprises: the device comprises a feature point extraction unit, a mapping point mapping unit, a mapping feature point searching unit, a translation rotation unit and a registration result output unit;
the feature point extraction unit is connected with the mapping point mapping unit, the mapping point mapping unit is connected with the mapping feature point searching unit, the mapping feature point searching unit is connected with the translation rotation unit, the translation rotation unit is connected with the feature point extraction unit and the registration result output unit, and the registration result output unit is connected with the feature point extraction unit;
The characteristic point extraction unit is used for selecting an ith characteristic point on the first blood vessel central line based on a selection principle, wherein i is more than or equal to 1 and less than or equal to X, and i is an integer;
the mapping point mapping unit is used for mapping the ith feature point onto a second blood vessel central line to obtain an ith mapping point;
the mapping feature point searching unit is used for searching points on the central line of the second blood vessel in a circle with the ith mapping point as a circle center and the number of the first designated pixel points as a radius by adopting the same selection principle as the ith feature point, and determining the searched points as the ith mapping feature point;
the translation rotation unit is used for selecting the (i+1) th feature point by adopting the feature point extraction unit, and searching by adopting the mapping point mapping unit and the mapping feature point searching unit to obtain the (i+1) th mapping feature point; translating or rotating the second blood vessel center line by adopting the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point until the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point meet a first preset standard, and stopping translating or rotating the current second blood vessel center line;
The registration result output unit is used for selecting the next feature point by adopting the feature point extraction unit, executing the mapping point mapping unit and the mapping feature point searching unit again until the spatial relationship similarity of the X feature points accords with a second preset standard, finishing registration of the first blood vessel center line and the current second blood vessel center line, and determining the current second blood vessel center line as a registration result.
Example 3:
the present embodiment proposes an electronic device including: the system comprises a memory and a processor, wherein the memory stores a computer program which is executed by the processor to execute the vessel center line registration method based on the magnetic resonance image.
The electronic device may be a mobile phone, a computer or a tablet computer, etc., comprising a memory and a processor, wherein the memory stores a computer program which, when executed by the processor, implements a method for registering a vessel centerline based on magnetic resonance images as described in the embodiments. It is to be appreciated that the electronic device can also include an input/output (I/O) interface, as well as a communication component.
Wherein the processor is configured to perform all or part of the steps in the magnetic resonance image based vessel centerline registration method as described in the above embodiments. The memory is used to store various types of data, which may include, for example, instructions for any application or method in the electronic device, as well as application-related data.
The processor may be an application specific integrated circuit (Application Specific Integrated Cricuit, ASIC), digital signal processor (Digital Signal Processor, DSP), programmable logic device (Programmable Logic Device, PLD), field programmable gate array (Field Programmable Gate Array, FPGA), controller, microcontroller, microprocessor or other electronic component implementation for performing the magnetic resonance image based vessel centerline registration method described in the above embodiments.
Example 4:
the present embodiment proposes a computer readable storage medium storing executable instructions that when executed cause a processor to perform the described magnetic resonance image based vessel centerline registration method.
If implemented as a software functional unit and sold or used as a stand-alone product, may be stored on a computer readable storage medium.
Based on such understanding, the technical solution of the present application may be embodied essentially or partly in the form of a software product stored in a storage medium, comprising instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method for registering a vessel centerline based on a magnetic resonance image according to the embodiments of the present application.
And the aforementioned storage medium includes: flash memory, hard disk, multimedia card, card memory (e.g., SD (Secure Digital Memory Card secure digital memory card) or DX (Memory Data Register, abbreviated as MDR, memory data register) memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, server, APP (Application, abbreviated as Application software) Application, etc., on which a computer program is stored, which when executed by a processor, implements the steps of the above-described magnetic resonance image-based vessel centerline registration method.
The various embodiments in this disclosure are described in a progressive manner, and identical and similar parts of the various embodiments are all referred to each other, and each embodiment is mainly described as different from other embodiments.
The scope of the present disclosure is not limited to the above-described embodiments, and it is apparent that various modifications and variations can be made to the present disclosure by those skilled in the art without departing from the scope and spirit of the disclosure. Such modifications and variations are intended to be included herein within the scope of the following claims and their equivalents.
Claims (10)
1. A method for vessel centerline registration based on magnetic resonance images, comprising:
step S1: acquiring N nuclear magnetic image sequences, wherein the N nuclear magnetic image sequences comprise 1 reference nuclear magnetic image sequence and N-1 nuclear magnetic image sequences to be registered;
step S2: carrying out global rigid registration on the N nuclear magnetic image sequences;
step S3: performing blood vessel segmentation on the N nuclear magnetic image sequences subjected to global rigid registration to obtain a first blood vessel segmentation result corresponding to the reference nuclear magnetic image sequence and N-1 second blood vessel segmentation results corresponding to the N-1 nuclear magnetic image sequences to be registered;
step S4: extracting a first vessel center line in a first vessel segmentation result;
step S5: extracting a second vessel center line in a second vessel segmentation result of the kth nuclear magnetic image sequence to be registered, wherein k is more than or equal to 2 and less than or equal to N-1;
step S6: selecting X characteristic points on the first blood vessel central line, searching X mapping characteristic points on the second blood vessel central line based on the X characteristic points, and registering the second blood vessel central line with the first blood vessel central line based on the X characteristic points and the X mapping characteristic points;
step S7: checking the X feature points, and under the condition that the checking is passed, determining that the vessel center lines of the current reference nuclear magnetic image sequence and the kth nuclear magnetic image sequence to be registered are successfully registered, wherein k=k+1, and executing the step S5 again until the vessel center lines of all N-1 nuclear magnetic image sequences to be registered are successfully registered, and finishing the vessel center line registration; if the verification is not passed, executing the step S6 again;
The selecting X feature points on the first vessel centerline, searching X mapping feature points on the second vessel centerline based on the X feature points, and registering the second vessel centerline with the first vessel centerline based on the X feature points and the X mapping feature points includes:
step S6.1: selecting an ith characteristic point on the first blood vessel central line based on a selection principle, wherein i is more than or equal to 1 and less than or equal to X, and i is an integer;
step S6.2: mapping the ith feature point to a second blood vessel central line to obtain an ith mapping point;
step S6.3: the same selection principle as the ith feature point is adopted, points on the central line of a second blood vessel are searched in circles taking the ith mapping point as the circle center and the number of the first designated pixel points as the radius, and the searched points are determined to be the ith mapping feature points;
step S6.4: selecting the i+1 th feature point by adopting the step S6.1, and searching by adopting the step S6.2 and the step S6.3 to obtain the i+1 th mapping feature point;
step S6.5: translating or rotating the second blood vessel center line by adopting the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point until the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point meet a first preset standard, and stopping translating or rotating the current second blood vessel center line;
Step S6.6: selecting the next feature point by adopting the step S6.1, and executing the steps S6.2-S6.5 again until the spatial relationship similarity of the X feature points meets a second preset standard, finishing the registration of the first blood vessel center line and the current second blood vessel center line, and determining the current second blood vessel center line as a registration result;
the selection principle of the characteristic points comprises the following steps: and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel bifurcation as a characteristic point, and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel curvature on the first blood vessel centerline as a characteristic point at which the blood vessel curvature is larger than a preset curvature threshold.
2. The method of claim 1, wherein the vessel bifurcation is a position where the vessel starts to extend in two different directions.
3. The method of claim 1, wherein the vessel curvature refers to a degree of curvature at each location on the vessel centerline.
4. The method for registering a vessel centerline based on a magnetic resonance image according to claim 1, wherein the searching for a point on a second vessel centerline in a circle with the i-th mapping point as a center and the first specified number of pixels as a radius using the same selection principle as the i-th mapping point, and determining the searched point as the i-th mapping feature point comprises:
Searching a point on the central line of the second blood vessel in a circle with the i-th mapping point as a circle center and the number of the first designated pixel points as a radius by adopting the same selection principle as the i-th characteristic point;
if a point located on the second vessel centerline can be searched, the searched point is used as an ith mapping feature point;
if the points on the second blood vessel central line cannot be searched, increasing the number of the specified pixel points, continuing to adopt the same selection principle as the ith feature point, searching the points on the second blood vessel central line in a circle with the ith mapping point as the circle center and the increased number of the specified pixel points as the radius until the points on the second blood vessel central line are searched, and taking the points obtained by searching at the moment as the ith mapping feature point.
5. The method for registering a vessel centerline based on a magnetic resonance image according to claim 1, wherein the spatial positional relationship between the i-th feature point and the i+1-th feature point and the spatial positional relationship between the i-th mapping feature point and the i+1-th mapping feature point conform to a first predetermined criterion, comprising: the Euclidean distance between the ith feature point and the (i+1) th feature point and between the ith mapping feature point and the (i+1) th mapping feature point is less than or equal to a first Euclidean distance threshold, and the cosine similarity between the ith feature point and the (i+1) th feature point and between the (i mapping feature point and the (i+1) th mapping feature point is less than or equal to a first cosine similarity threshold.
6. The method for registering a vessel centerline based on a magnetic resonance image according to claim 1, wherein the rule of value of X comprises:if the number of the mapping feature points required to be obtained on the central line of each second blood vessel in the nuclear magnetic image sequence to be registered is not less thanpIf the value rule of X is thatWherein->And extracting a blood vessel center line from each blood vessel in the reference nuclear magnetic image sequence and the nuclear magnetic image sequence to be registered for the number of blood vessels contained in the kth nuclear magnetic image sequence to be registered.
7. The method of magnetic resonance image based vessel centerline registration as set forth in claim 1, wherein the verifying the X feature points comprises:
reconstructing a cross section of a blood vessel where the current first blood vessel center line is located at the ith characteristic point;
in the cross section, taking the ith characteristic point as a circle center and taking the number of second designated pixel points as a radius to make a circle;
performing similarity calculation on gray values of the reference nuclear magnetic image sequence and the nuclear magnetic image sequence to be registered in the circle by adopting a perception hash algorithm to obtain hash values of the reference nuclear magnetic image sequence and hash values of the nuclear magnetic image sequence to be registered at positions corresponding to all pixels in the circle;
Comparing the hash value of the reference nuclear magnetic image sequence at the corresponding positions of all pixels in the circle with the hash value of the nuclear magnetic image sequence to be registered; and if the number of the hash values of the corresponding positions is not equal to or less than the first similarity threshold, checking passing, otherwise, checking failing.
8. A vessel centerline registration device based on magnetic resonance images, comprising:
the image acquisition module is used for acquiring N nuclear magnetic image sequences, wherein the N nuclear magnetic image sequences comprise 1 reference nuclear magnetic image sequence and N-1 nuclear magnetic image sequences to be registered;
the global registration module is used for carrying out global rigid registration on the N nuclear magnetic image sequences;
the vessel segmentation module is used for carrying out vessel segmentation on the N nuclear magnetic image sequences subjected to global rigid registration to obtain a first vessel segmentation result corresponding to the reference nuclear magnetic image sequence and a second vessel segmentation result corresponding to the N-1 nuclear magnetic image sequences to be registered;
the first central line extraction module is used for extracting a first blood vessel central line in the first blood vessel segmentation result;
the second central line extraction module is used for extracting a second blood vessel central line in a second blood vessel segmentation result of the kth nuclear magnetic image sequence to be registered, wherein k is more than or equal to 2 and less than or equal to N-1;
The local registration module is used for selecting X characteristic points on the first blood vessel central line, searching X mapping characteristic points on the second blood vessel central line based on the X characteristic points, and registering the second blood vessel central line with the first blood vessel central line based on the X characteristic points and the X mapping characteristic points;
the verification module is used for verifying the X feature points, under the condition that verification is passed, determining that the vessel center lines of the current reference nuclear magnetic image sequence and the kth nuclear magnetic image sequence to be registered are successfully registered, wherein k=k+1, and executing the second center line extraction module again until the vessel center lines of all N-1 nuclear magnetic image sequences to be registered are successfully registered, and finishing the vessel center line registration; executing the registration module again in case the verification is not passed;
the local registration module comprises: the device comprises a feature point extraction unit, a mapping point mapping unit, a mapping feature point searching unit, a translation rotation unit and a registration result output unit;
the characteristic point extraction unit is used for selecting an ith characteristic point on the first blood vessel central line based on a selection principle, wherein i is more than or equal to 1 and less than or equal to X, and i is an integer;
the mapping point mapping unit is used for mapping the ith feature point onto a second blood vessel central line to obtain an ith mapping point;
The mapping feature point searching unit is used for searching points on the central line of the second blood vessel in a circle with the ith mapping point as a circle center and the number of the first designated pixel points as a radius by adopting the same selection principle as the ith feature point, and determining the searched points as the ith mapping feature point;
the translation rotation unit is used for selecting the (i+1) th feature point by adopting the feature point extraction unit, and searching by adopting the mapping point mapping unit and the mapping feature point searching unit to obtain the (i+1) th mapping feature point; translating or rotating the second blood vessel center line by adopting the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point until the spatial position relation between the ith characteristic point and the (i+1) th characteristic point and the spatial position relation between the ith mapping characteristic point and the (i+1) th mapping characteristic point meet a first preset standard, and stopping translating or rotating the current second blood vessel center line;
the registration result output unit is used for selecting the next feature point by adopting the feature point extraction unit, and executing the mapping point mapping unit and the mapping feature point searching unit again until the spatial relationship similarity of X feature points accords with a second preset standard, and the registration of the first vessel center line and the current second vessel center line is finished, and the current second vessel center line is determined as a registration result;
The selection principle of the characteristic points comprises the following steps: and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel bifurcation as a characteristic point, and if the blood vessel bifurcation is not present on the first blood vessel centerline, selecting the blood vessel curvature on the first blood vessel centerline as a characteristic point at which the blood vessel curvature is larger than a preset curvature threshold.
9. An electronic device, comprising: a memory and a processor, the memory having stored thereon a computer program which, when executed by the processor, performs the magnetic resonance image-based vessel centerline registration method of any one of claims 1 to 7.
10. A computer readable storage medium storing executable instructions that when executed cause a processor to perform the magnetic resonance image based vessel centerline registration method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310935214.7A CN116664644B (en) | 2023-07-28 | 2023-07-28 | Vessel center line registration method and device based on magnetic resonance image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310935214.7A CN116664644B (en) | 2023-07-28 | 2023-07-28 | Vessel center line registration method and device based on magnetic resonance image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116664644A CN116664644A (en) | 2023-08-29 |
| CN116664644B true CN116664644B (en) | 2023-11-28 |
Family
ID=87715649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310935214.7A Active CN116664644B (en) | 2023-07-28 | 2023-07-28 | Vessel center line registration method and device based on magnetic resonance image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116664644B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451983A (en) * | 2017-07-18 | 2017-12-08 | 中山大学附属第六医院 | The three-dimensional fusion method and system of CT images |
| CN112164020A (en) * | 2020-03-31 | 2021-01-01 | 苏州润迈德医疗科技有限公司 | Method, device, analysis system and storage medium for accurately extracting blood vessel center line |
| CN113298831A (en) * | 2021-06-30 | 2021-08-24 | 上海联影医疗科技股份有限公司 | Image segmentation method and device, electronic equipment and storage medium |
| WO2021208739A1 (en) * | 2020-11-25 | 2021-10-21 | 平安科技（深圳）有限公司 | Method and apparatus for evaluating blood vessel in fundus color image, and computer device and medium |
| CN115359077A (en) * | 2022-05-27 | 2022-11-18 | 天津远景科技服务有限公司 | Method and device for determining black blood vessel model |
| CN115861189A (en) * | 2022-11-16 | 2023-03-28 | 深圳睿心智能医疗科技有限公司 | Image registration method and device, electronic equipment and storage medium |
-
2023
- 2023-07-28 CN CN202310935214.7A patent/CN116664644B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451983A (en) * | 2017-07-18 | 2017-12-08 | 中山大学附属第六医院 | The three-dimensional fusion method and system of CT images |
| CN112164020A (en) * | 2020-03-31 | 2021-01-01 | 苏州润迈德医疗科技有限公司 | Method, device, analysis system and storage medium for accurately extracting blood vessel center line |
| WO2021208739A1 (en) * | 2020-11-25 | 2021-10-21 | 平安科技（深圳）有限公司 | Method and apparatus for evaluating blood vessel in fundus color image, and computer device and medium |
| CN113298831A (en) * | 2021-06-30 | 2021-08-24 | 上海联影医疗科技股份有限公司 | Image segmentation method and device, electronic equipment and storage medium |
| CN115359077A (en) * | 2022-05-27 | 2022-11-18 | 天津远景科技服务有限公司 | Method and device for determining black blood vessel model |
| CN115861189A (en) * | 2022-11-16 | 2023-03-28 | 深圳睿心智能医疗科技有限公司 | Image registration method and device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 应用于造影图像的血管中心线全自动提取方法;肖若秀，杨健，刘越，王涌天;《计算机辅助设计与图形学学报》;第25卷(第3期);第1-9页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116664644A (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7787683B2 (en) | Tree structure based 2D to 3D registration | |
| CN111028205B (en) | Eye pupil positioning method and device based on binocular distance measurement | |
| CN109640066B (en) | Method and device for generating high-precision dense depth image | |
| CN113298831B (en) | Image segmentation method and device, electronic equipment and storage medium | |
| Fang et al. | Elastic registration for retinal images based on reconstructed vascular trees | |
| CN109767448B (en) | Segmentation model training method and device | |
| CN111105452A (en) | High-low resolution fusion stereo matching method based on binocular vision | |
| CN114219717A (en) | Point cloud registration method and device, electronic equipment and storage medium | |
| CN106548491A (en) | A kind of method for registering images, its image interfusion method and its device | |
| CN109816650A (en) | A kind of target area recognition methods and its system based on two-dimentional DSA image | |
| Tang et al. | Retinal image registration based on robust non-rigid point matching method | |
| KR20180088061A (en) | Method and apparatus for registering a CT image with a surface scan model | |
| CN116664644B (en) | Vessel center line registration method and device based on magnetic resonance image | |
| CN113299385A (en) | Pancreatic cystic lesion clinical decision method and system based on deep learning | |
| Li et al. | A Fast Subpixel Registration Algorithm Based on Single‐Step DFT Combined with Phase Correlation Constraint in Multimodality Brain Image | |
| CN111681237B (en) | Image registration method, device, computer equipment and storage medium | |
| CN113379770B (en) | Construction method of nasopharyngeal carcinoma MR image segmentation network, image segmentation method and device | |
| CN111739004B (en) | Image processing method, device and storage medium | |
| CN114170258A (en) | Image segmentation method and device, electronic equipment and storage medium | |
| CN114119688A (en) | Single-mode medical image registration method before and after coronary angiography based on deep learning | |
| CN112288639A (en) | Image contrast subtraction method and device, terminal equipment and storage medium | |
| CN114331942A (en) | Method, system, computer storage medium and device for processing multi-stage medical images | |
| US8363914B2 (en) | Method and apparatus for processing medical imaging data using phase information | |
| Pan et al. | A hybrid optimization method for maximum mutual information registration | |
| CN116228779B (en) | Blood vessel center path extraction method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |