Image registration method, system and storage medium for a vascular intervention procedure

Abstract

The present application provides a kind of vascular intervention surgery image registration method, comprising the following steps: obtaining reference image and target image;By extracting the corresponding multiple sets of feature points on the catheter of reference image and the catheter of target image, and according to the corresponding relationship between the coordinates of multiple sets of feature points, the transformation parameters of reference image to target image are calculated, finally, according to the transformation parameters, the reference image is registered with target image, so that the catheter in the reference image can be accurately superimposed on the catheter of target image, so as to realize accurate splicing of the blood vessels in the reference image and the catheter in the target image;When the position of catheter in intraoperative real-time DSA image is offset, it is not necessary to re-inject contrast medium to obtain accurate blood vessel boundary image, which greatly reduces the intake of contrast medium and reduces the harm of contrast medium to patients.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical image processing, and in particular to an image registration method and system for vascular intervention surgery and a storage medium. BACKGROUND

[0002] The rapid development of science and technology has brought great changes to people's life, and has promoted the progress of society, and the medical field is no exception. The progress of computer technology, image processing technology and medical physics provides more and more powerful tools for medical diagnosis and treatment.

[0003] Cardiovascular disease is one of the major problems threatening human health in the world today, and both the incidence and mortality rate are in the forefront of various diseases, so preventing and treating cardiovascular disease is of great significance to people's life and health. By segmenting the coronary artery on the medical image and superimposing the blood vessel boundary in real time on the DSA image during the intervention surgery, the position of the guide wire and catheter in the blood vessel can be observed more intuitively and effectively, thereby reducing the injection of contrast agent during the operation and reducing the harm of contrast agent to the patient.

[0004] During the vascular intervention surgery, contrast agent needs to be injected into the patient's body to make the DSA device display the blood vessel profile, otherwise only the current position of the catheter (a surgical instrument) can be displayed. The doctor judges the next operation method by observing the blood vessel pattern at the injection moment, but the contrast agent is nephrotoxic to the patient, and frequent use has a large load on the liver and kidney function of the patient, so by recording a part of the blood vessel visualization video, the position of the catheter in the stored blood vessel visualization part is extracted by processing, and by comparing with the real-time image frame without injection of contrast agent (only the catheter can be displayed), the pre-stored blood vessel image is superimposed on the catheter of the image frame without injection of contrast agent to be displayed in real time. However, due to respiration, heartbeat and the like, the pre-stored catheter image is not completely the same as the real-time catheter image, resulting in poor accuracy of the vascular intervention surgery. Therefore, an image registration method for vascular intervention surgery is needed to register the catheter, so that the blood vessels are superimposed on the correct position after transformation. SUMMARY

[0005] The present application provides an image registration method and system for vascular intervention surgery and a storage medium, which solves the problem of poor accuracy of vascular intervention surgery caused by the fact that the pre-stored catheter image is not completely the same as the real-time catheter image in the prior art.

[0006] The technical solution of the present application is as follows:

[0007] According to one aspect of the present application, an image registration method for vascular intervention surgery is provided, comprising the following steps:

[0008] Obtaining a reference image and a target image; wherein the reference image is an image including a catheter and blood vessels obtained by a DSA device after injecting contrast agent, and the target image is a real-time image of the catheter obtained by the DSA device without injecting the contrast agent;

[0009] Extracting a plurality of groups of corresponding feature points on the catheter of the reference image and the target image;

[0010] Calculating transformation parameters from the reference image to the target image according to the corresponding relationship between the coordinates of the plurality of groups of feature points;

[0011] Registering the reference image and the target image according to the transformation parameters.

[0012] As a preferred scheme of the present application, the HOG similar point detection algorithm is used to extract the corresponding feature points of the reference image and the target image.

[0013] As a preferred scheme of the present application, the method for calculating the transformation parameters from the reference image to the target image is as follows:

[0014] Supposing that the feature point set on the reference image is {A1, A2, …, An}, and the corresponding feature point set on the target image is {B1, B2, …, Bn};

[0015] Connecting the plurality of feature points on the reference image two by two to obtain a feature line segment set {L1, L2, …, Lm} on the reference image, and connecting the plurality of feature points on the target image two by two to obtain a feature line segment set {L’1, L’2, …, L’n} on the target image; m}; m

[0016] Taking any one group of corresponding feature line segments L k and L’ k , k∈1,2,…,m; L k The coordinates of the two end points Ai and Aj are: Ai(Ai.x, Ai.y), Aj(Aj.x, Aj.y); L’ k The coordinates of the two end points Bi and Bj are: Bi(Bi.x, Bi.y), Bj(Bj.x, Bj.y); i,j∈1,2,…,n;

[0017] If the matrix is a non-singular matrix, then

[0018]

[0019] Wherein, F k is the transformation parameter between the feature line segment L k and L’ k .

[0020] If the matrix ​If it is a singular matrix, then remove the characteristic line segment L. k and L' k ;

[0021] Traverse the feature line segment set {L1,L2,…,L m} and {L'1,L'2,…,L' m The transformation parameter set {F1, F2, ..., F} is calculated using the above method. n}, n≤m, select appropriate transformation parameters F according to the transformation parameter set. * As transformation parameters from the reference image to the target image.

[0022] As a preferred embodiment of the present invention, the transformation parameters include scaling ratio, rotation angle, and translation amount; a suitable transformation parameter F is selected according to the transformation parameter set. * The method is as follows: from the transformation parameter set {F1,F2,…,F…} n In the transformation parameters F, the median of the scaling factor, the median of the rotation angle, and the median of the translation are selected respectively. * .

[0023] According to another aspect of the present invention, an image registration system for vascular interventional surgery is provided, comprising:

[0024] The image acquisition module is used to acquire a reference image and a target image; wherein, the reference image is an image including the catheter and blood vessel acquired by the DSA device after the injection of contrast agent, and the target image is a catheter image acquired by the DSA device without the injection of contrast agent;

[0025] The feature point extraction module is used to extract multiple sets of feature points corresponding to the conduits in the reference image and the conduits in the target image;

[0026] The calculation module is used to calculate the transformation parameters from the reference image to the target image based on the correspondence between the coordinates of multiple sets of feature points;

[0027] The registration module is used to register the reference image and the target image according to the transformation parameters.

[0028] According to another aspect of the present invention, a storage medium is provided that stores a computer program, which, when executed by a processor, implements the steps of the above-described registration method.

[0029] Beneficial effects

[0030] Compared with the prior art, the present application has the beneficial effects that: the present application extracts corresponding multiple sets of feature points on the catheter of the reference image and the catheter of the target image, calculates the transformation parameters of the reference image to the target image according to the corresponding relationship between the coordinates of the multiple sets of feature points, and finally registers the reference image and the target image according to the transformation parameters, so that the catheter in the reference image can be accurately superimposed on the catheter of the target image, thereby realizing accurate splicing of the blood vessels in the reference image and the catheter in the target image; when the position of the catheter in the real-time DSA image during the operation is offset, there is no need to re-inject contrast medium to obtain an accurate blood vessel boundary image, which greatly reduces the intake of contrast medium and reduces the harm of contrast medium to patients. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0032] Figure 1 The flowchart of the image registration method of the present application for vascular interventional surgery;

[0033] Figure 2 The schematic diagram of the reference image and the target image obtained in the embodiment of the present application;

[0034] Figure 3 The schematic diagram of the feature points on the reference image being connected and the feature points on the target image being connected in the embodiment of the present application;

[0035] Figure 4 The schematic diagram of the feature line segments selected from the reference image and the target image in the embodiment of the present application;

[0036] Figure 5 The schematic diagram of the reference image being registered with the real-time target image in the embodiment of the present application. DETAILED DESCRIPTION

[0037] The technical solutions of the present application will be described in detail below with reference to the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0038] Referring to Figure 1 The present embodiment provides an image registration method for vascular interventional surgery, comprising the following steps:

[0039] As shown in Figure 2 , a reference image and a target image are acquired; wherein, Figure 2 the reference image on the left is an image including a catheter and a blood vessel acquired by a DSA device after injecting contrast agent, wherein the edge of the catheter is labeled with a white contour line, Figure 2 the target image on the right is a real-time image of the catheter acquired by the DSA device without injecting contrast agent;

[0040] a plurality of sets of feature points corresponding to the catheter of the reference image and the catheter of the target image are extracted;

[0041] a transformation parameter from the reference image to the target image is calculated according to the corresponding relationship between the coordinates of the plurality of sets of feature points;

[0042] the reference image is registered with the target image according to the transformation parameter.

[0043] As a preferred scheme of the embodiment, a HOG similar point detection algorithm is used to extract the corresponding feature points of the reference image and the target image, and the specific method is that the acquired reference image and target image are preprocessed, including:

[0044] the image is converted into a grayscale image;

[0045] a Gamma correction method is used to standardize the color space of the image, the purpose being to adjust the contrast of the image and reduce the influence of local shadows and illumination changes of the image, while suppressing the interference of noise;

[0046] a Gaussian function is used to smooth the image to remove noise points;

[0047] the preprocessed reference image and target image are input into a pre-trained feature point recognition model, so as to extract the feature points of the reference image and the target image.

[0048] As a preferred scheme of the embodiment, the method for calculating the transformation parameter from the reference image to the target image is:

[0049] assuming that the feature point set on the reference image is {A1, A2, A3, A4, A5} and the corresponding feature point set on the target image is {B1, B2, B3, B4, B5};

[0050] As shown in Figure 3 , a plurality of feature points on the reference image are connected two by two to obtain a feature line segment set {L1, L2, …, L 10} and a plurality of feature points on the target image are connected two by two to obtain a feature line segment set {L’1, L’2, …, L’ 10};

[0051] AsFigure 4 As shown, take any one set of corresponding feature line segments L k and L’ k , k∈1, 2, …, 10; L k The coordinates of the two endpoints A1, A2 are: A1(A1.x, A1.y), A2(A2.x, A2.y); L’ k The coordinates of the two endpoints B1, B2 are: B1(B1.x, B1.y), B2(B2.x, B2.y);

[0052] Suppose the upper left corner of the two images is the coordinate origin, then the transformation relationship of the feature line segment A1-A2 in the coordinate system of the reference image to the coordinate system of the feature line segment B1-B2 in the target image can be calculated as:

[0053]

[0054] If the matrix is a non-singular matrix, then

[0055]

[0056] Where, F k is the transformation parameter between the feature line segment L k and L’ k ;

[0057] If the matrix is a singular matrix, then the feature line segment L k and L’ k are removed;

[0058] Iterate the feature line segment sets {L1, L2, …, L 10} and {L’1, L’2, …, L’ 10}, and calculate the transformation parameter set {F1, F2, …, F n} using the above method, n≤10, and select a suitable transformation parameter F * from the transformation parameter set as the transformation parameter of the reference image to the target image.

[0059] As a preferred scheme of the embodiment, the transformation parameter includes a scaling ratio, a rotation angle, and a translation amount; and the method for selecting a suitable transformation parameter F * from the transformation parameter set is: selecting the median of the scaling ratio, the median of the rotation angle, and the median of the translation amount from the transformation parameter set {F1, F2, …, F n} respectively to form a new transformation parameter F * , and using the transformation parameter F *That is, the reference image can be registered with the real-time target image, and the catheter in the reference image can be superimposed on the catheter in the target image, so that the blood vessels in the reference image can be accurately spliced with the catheter in the target image, thereby improving the operation precision without repeated injection of contrast agent. The registered image is as shown in Figure 5 .

[0060] Corresponding to the above registration method, the embodiment also provides an image registration system for a vascular interventional operation, comprising:

[0061] an image acquisition module, configured to acquire a reference image and a target image; wherein the reference image is an image including a catheter and blood vessels acquired by a DSA device after injection of contrast agent, and the target image is a catheter image acquired by the DSA device without injection of contrast agent;

[0062] a feature point extraction module, configured to extract a plurality of groups of corresponding feature points on the catheter of the reference image and the catheter of the target image;

[0063] a calculation module, configured to calculate a transformation parameter of the reference image to the target image according to the corresponding relationship between the coordinates of the plurality of groups of feature points;

[0064] a registration module, configured to register the reference image with the target image according to the transformation parameter.

[0065] Corresponding to the above registration method, the embodiment also provides a storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the above registration method.

[0066] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An image registration method for vascular interventional surgery, characterized in that, Includes the following steps: Acquire a reference image and a target image; wherein, the reference image is an image including the catheter and blood vessel acquired by the DSA device after the injection of contrast agent, and the target image is a real-time image of the catheter acquired by the DSA device without the injection of contrast agent; Extract multiple sets of feature points corresponding to the ducts in the reference image and the ducts in the target image; The transformation parameters from the reference image to the target image are calculated based on the correspondence between the coordinates of multiple sets of feature points. The method is as follows: Assume the feature point set on the reference image is {A1,A2,…,An}, and the corresponding feature point set on the target image is {B1,B2,…,Bn}; Connecting each pair of feature points on the reference image yields a feature line segment set {L1, L2, ..., L...} m Connecting each pair of feature points on the target image yields a feature line segment set {L'1, L'2, ..., L'}. m }; Take any set of corresponding feature line segments L k and L' k , k∈1,2,…,m;L k The coordinates of the two endpoints Ai and Aj are: Ai(Ai.x, Ai.y), Aj(Aj.x, Aj.y); L' k The coordinates of the two endpoints Bi and Bj are: Bi (Bi.x, Bi.y), Bj (Bj.x, Bj.y); i,j∈1,2,…,n; If matrix If it is a non-singular matrix, then ； Among them, F k Characteristic line segment L k With L' k Transformation parameters between; If matrix If it is a singular matrix, then remove the characteristic line segment L. k and L' k ; Traverse the feature line segment set {L1,L2,…,L m } and {L'1,L'2,…,L' m The transformation parameter set {F1, F2, ..., F} is calculated using the above method. n }, n≤m, select appropriate transformation parameters F according to the transformation parameter set. * As transformation parameters from the reference image to the target image; The reference image and the target image are registered according to the transformation parameters.

2. The image registration method for vascular interventional surgery as described in claim 1, characterized in that, The HOG similarity point detection algorithm is used to extract feature points corresponding to the reference image and the target image.

3. The image registration method for vascular interventional surgery as described in claim 1, characterized in that, The transformation parameters include scaling ratio, rotation angle, and translation amount; a suitable transformation parameter F is selected according to the transformation parameter set. * The method is as follows: from the transformation parameter set {F1,F2,…,F…} n In the transformation parameters F, the median of the scaling factor, the median of the rotation angle, and the median of the translation are selected respectively. * .

4. An image registration system for vascular interventional surgery, used to implement the image registration method for vascular interventional surgery as described in any one of claims 1 to 3, characterized in that, include: The image acquisition module is used to acquire a reference image and a target image; wherein, the reference image is an image including the catheter and blood vessel acquired by the DSA device after the injection of contrast agent, and the target image is a catheter image acquired by the DSA device without the injection of contrast agent; The feature point extraction module is used to extract multiple sets of feature points corresponding to the conduits in the reference image and the conduits in the target image; The calculation module is used to calculate the transformation parameters from the reference image to the target image based on the correspondence between the coordinates of multiple sets of feature points; The registration module is used to register the reference image and the target image according to the transformation parameters.

5. A storage medium, characterized in that, The storage medium stores a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1 to 3.

Images