A phantom-based system, method, equipment, and media for detecting calcified plaques in coronary CTA.

Abstract

This invention provides a coronary CTA calcified plaque detection system, method, device, and medium based on a phantom, relating to the field of applied medical technology. In this embodiment, after adjusting the CT equipment, at least one CTA image of all plaque phantoms can be acquired. After acquiring the patient's coronary CTA, the area adjacent to the plaque location is analyzed by computer. The analysis results are then matched with the phantom image analysis results using the same algorithm and designed indicators to select the phantom that best simulates the plaque situation in the current CTA. This phantom information is used to assist in reconstructing the true condition of the plaque on the CTA. Therefore, this embodiment of the invention can use plaque phantoms to match CTA to assist in reconstructing the true size of coronary plaques on CTA, increasing the reliability of non-invasive coronary CTA in plaque analysis.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of applied medical technology, and in particular to a coronary CTA calcified plaque detection system, method, device and medium based on a phantom. Background Technology

[0002] Coronary artery disease (CAD) is a common chronic disease among middle-aged and elderly people worldwide, seriously endangering human life. The pathogenesis of CAD is the deposition of lipids and other blood components in the arterial intima, smooth muscle cell proliferation, and increased collagen fibers, leading to atherosclerotic, lipid-containing necrotic lesions and hardening of the blood vessel wall. These pathological changes in the vascular intima result in plaque formation, causing narrowing of the blood vessels, affecting blood flow to the myocardium, and causing myocardial ischemia. When plaques detach acutely, they can cause acute coronary syndrome, seriously endangering life. Therefore, accurate detection and assessment of the size, shape, and composition of coronary plaques are of great significance for the treatment of CAD.

[0003] Coronary CT angiography (CTA) is one of the most commonly used clinical examinations for coronary artery disease. It only requires the patient to receive an intravenous injection of contrast agent, followed by a high-speed CT scan of the heart, providing three-dimensional images of the coronary arteries. It can clearly display the geometry of the coronary arteries. However, when there is severe plaque calcification, CT imaging magnifies the calcified plaques and produces radial low-density artifacts, making it impossible to accurately assess the size and shape of the plaques. This leads to doctors being unable to correctly determine the degree of coronary artery stenosis, affecting the formulation of subsequent treatment plans.

[0004] Therefore, there is an urgent need for a new coronary CTA method for detecting calcified plaques. Summary of the Invention

[0005] This invention provides a coronary CTA calcified plaque detection system, method, device, and medium based on a phantom, to at least partially solve the problems existing in related technologies.

[0006] A first aspect of this invention provides a coronary CTA calcified plaque detection system based on a phantom, the system comprising:

[0007] The phantom CTA image acquisition module is used to acquire CTA images of multiple patch phantoms;

[0008] The coronary artery CTA image acquisition module is used to acquire the CTA image to be detected;

[0009] The matching module is used to match the CTA image of the patch phantom and the CTA image to be detected using a pre-trained matching model algorithm, so as to obtain the target phantom that matches the CTA image to be detected;

[0010] The restoration module is used to restore the size of the calcified plaques in the CTA image to be detected based on the target phantom.

[0011] Optionally, the restoration module is specifically used for:

[0012] The image information of the CTA image of the target phantom is obtained and restored by digital image processing to obtain the true size of the calcified plaques in the CTA image. The image information includes one or more of the following: phantom point spread function, phantom artifact proportion, and phantom artifact contour.

[0013] Optionally, the restoration module is further configured to: determine the true size of the patch in the CTA image to be detected by using a pre-trained restoration model and the CTA image of the target phantom;

[0014] The domain of the restoration algorithm F is the CTA image and the labeled data, and the range is a set G consisting of the initial model and all its parameter combination models. The element g∈G is a function whose domain and range are both CTA image matrices, F(a,b)=g, g(c)=d, where a∈A, b∈B, c is the CTA to be detected, d is the restored CTA, A is the CTA image, and B is the labeled data.

[0015] Optionally, the restoration model is trained according to the following steps:

[0016] A plaque phantom is manufactured, and the phantom information of the plaque phantom is used as sample phantom annotation data. The sample phantom annotation data includes: phantom form, calcification ratio, plaque size, and plaque type.

[0017] Obtain CTA images of each patch phantom as sample CTA images;

[0018] The preset model is trained using the sample CTA image and the sample phantom annotation data to obtain the restored model.

[0019] Optionally, the matching module specifically includes:

[0020] The analysis submodule is used to analyze the CTA images of multiple patch phantoms and the CTA image to be detected using analysis algorithms to obtain the feature vector combination matrix corresponding to the CTA images of multiple patch phantoms and the feature vector combination matrix of the CTA image to be detected.

[0021] The combination submodule is used to combine the feature vector combination matrix corresponding to the CTA image of each patch body model with the feature vector combination matrix of the CTA image to be detected, respectively, to obtain multiple element pairs;

[0022] The evaluation submodule is used to evaluate the similarity of each element pair using an index evaluation function, obtain an evaluation score for each element pair, and select the patch phantom corresponding to the element pair with the highest evaluation score as the target phantom that matches the CTA image to be detected.

[0023] Optionally, the evaluation submodule is specifically used for:

[0024] By using multiple evaluation functions, the similarity of each element pair is evaluated, and the score of each element pair under different evaluation functions is obtained.

[0025] The scores of each element pair under different indicator evaluation functions are weighted and calculated to obtain the comprehensive evaluation score of each element pair;

[0026] The patch phantom corresponding to the element with the highest comprehensive evaluation score is used as the target phantom that matches the CTA image to be detected.

[0027] Optionally, the CTA image of the plaque phantom and the CTA image to be detected are CTA images taken by the same device under the same conditions.

[0028] A second aspect of this invention provides a method for detecting coronary CTA calcified plaques based on a phantom, the method comprising:

[0029] Step 1: Obtain CTA images of multiple patch phantoms;

[0030] Step 2: Acquire the CTA image to be detected;

[0031] Step 3: Using a pre-trained matching model algorithm, match the CTA image of the patch phantom with the CTA image to be detected to obtain the target phantom that matches the CTA image to be detected;

[0032] Step 4: Reconstruct the size of the calcified plaques in the CTA image to be detected based on the target phantom.

[0033] Optionally, step 4 includes:

[0034] The image information of the CTA image of the target phantom is obtained and restored by digital image processing to obtain the true size of the calcified plaques in the CTA image. The image information includes one or more of the following: phantom point spread function, phantom artifact proportion, and phantom artifact contour.

[0035] Optionally, step 4 further includes: determining the true size of the patch in the CTA image to be detected by using a pre-trained reconstruction model and the CTA image of the target phantom;

[0036] The domain of the restoration algorithm F is the CTA image and the labeled data, and the range is a set G consisting of the initial model and all its parameter combination models. The element g∈G is a function whose domain and range are both CTA image matrices, F(a,b)=g, g(c)=d, where a∈A, b∈B, c is the CTA to be detected, d is the restored CTA, A is the CTA image, and B is the labeled data.

[0037] Optionally, the restoration model is trained according to the following steps:

[0038] A plaque phantom is manufactured, and the phantom information of the plaque phantom is used as sample phantom annotation data. The sample phantom annotation data includes: phantom form, calcification ratio, plaque size, and plaque type.

[0039] Obtain CTA images of each patch phantom as sample CTA images;

[0040] The preset model is trained using the sample CTA image and the sample phantom annotation data to obtain the restored model.

[0041] Optionally, step 3 includes:

[0042] The CTA images of multiple patch phantoms and the CTA image to be detected are analyzed using an analysis algorithm to obtain the feature vector combination matrix corresponding to the CTA images of multiple patch phantoms and the feature vector combination matrix of the CTA image to be detected.

[0043] The feature vector combination matrix corresponding to the CTA image of each patch body model is combined with the feature vector combination matrix of the CTA image to be detected to obtain multiple element pairs;

[0044] The similarity of each element pair is evaluated using an index evaluation function, and an evaluation score is obtained for each element pair. The patch phantom corresponding to the element pair with the highest evaluation score is used as the target phantom that matches the CTA image to be detected.

[0045] Optionally, the similarity of each element pair is evaluated using an index evaluation function to obtain an evaluation score for each element pair. The patch phantom corresponding to the element pair with the highest evaluation score is used as the target phantom that matches the CTA image to be detected, including:

[0046] By using multiple evaluation functions, the similarity of each element pair is evaluated, and the score of each element pair under different evaluation functions is obtained.

[0047] The scores of each element pair under different indicator evaluation functions are weighted and calculated to obtain the comprehensive evaluation score of each element pair;

[0048] The patch phantom corresponding to the element with the highest comprehensive evaluation score is used as the target phantom that matches the CTA image to be detected.

[0049] Optionally, the CTA image of the plaque phantom and the CTA image to be detected are CTA images taken by the same device under the same conditions.

[0050] A third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the steps of the coronary CTA calcified plaque detection method based on a phantom as described in the second aspect of the present invention.

[0051] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the coronary CTA calcified plaque detection method based on a phantom as described in the second aspect of the present invention.

[0052] In this embodiment of the invention, a calcified plaque phantom can be designed, and the phantom information combined with computer technology can be used to analyze the plaque information in the coronary CTA image to be detected.

[0053] Specifically, in this embodiment of the invention, after adjusting the CT equipment, at least one CTA image of all plaque phantoms can be acquired. After acquiring the patient's coronary CTA for examination, the area adjacent to the plaque location is analyzed by computer. The methods used in the analysis include: grayscale statistics, grayscale analysis, feature extraction (various filtering features, etc.), and plaque information prediction methods based on machine learning (such as predicting the calcification ratio and size). The analysis results are matched with the phantom images under the same algorithm using designed indicators to select the phantom that best simulates the plaque situation in the current CTA. This phantom information is used to assist in reconstructing the true condition of the plaque on the CTA. Therefore, this embodiment of the invention can use plaque phantoms to match CTA to assist in reconstructing the true size of coronary plaques on CTA, increasing the reliability of non-invasive coronary CTA in plaque analysis. Attached Figure Description

[0054] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1This is a structural block diagram of a coronary CTA calcified plaque detection device based on a phantom according to an embodiment of the present invention;

[0056] Figure 2 This is a flowchart of a coronary CTA calcified plaque detection method based on a phantom, according to an embodiment of the present invention. Detailed Implementation

[0057] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] Reference Figure 1 Figure 1 shows a structural block diagram of a coronary CTA calcified plaque detection system based on a phantom according to an embodiment of the present invention. As shown in the figure, the coronary CTA calcified plaque detection system based on a phantom provided by this embodiment of the present invention includes:

[0059] The phantom CTA image acquisition module 101 is used to acquire CTA images of multiple patch phantoms.

[0060] The coronary artery CTA image acquisition module 102 is used to acquire the CTA image to be detected.

[0061] The matching module 103 is used to match the CTA image of the patch phantom and the CTA image to be detected using a pre-trained matching model algorithm, so as to obtain a target phantom that matches the CTA image to be detected.

[0062] The restoration module 104 is used to restore the size of the calcified plaques in the CTA image to be detected based on the target phantom.

[0063] In this embodiment of the invention, a calcified plaque phantom can be designed, and the phantom information combined with computer technology can be used to analyze the plaque information in the coronary CTA image to be detected.

[0064] Specifically, in this embodiment of the invention, multiple phantoms simulating the internal environment of the human body can be designed using different phantom parameters. These parameters include at least: phantom form (simulating the entire thoracic cavity or blood vessels, etc.), calcification ratio, plaque size, and plaque type. Different phantom parameters can be combined to form plaque phantoms corresponding to different parameter combinations.

[0065] In this embodiment of the invention, after adjusting the CT equipment, at least one CTA image of all plaque phantoms can be acquired. After acquiring the patient's coronary CTA for examination, the area adjacent to the plaque location is analyzed by computer. The methods used in the analysis include: grayscale statistics, grayscale analysis, feature extraction (various filtering features, etc.), and plaque information prediction methods based on machine learning (such as predicting the calcification ratio and size). The analysis results are matched with the phantom images using the same algorithm and designed indicators to select the phantom that best simulates the plaque situation in the current CTA. This phantom information is used to assist in reconstructing the true condition of the plaque on the CTA. Therefore, this embodiment of the invention can use plaque phantoms to match CTA to assist in reconstructing the true size of coronary plaques on CTA, increasing the reliability of non-invasive coronary CTA in plaque analysis.

[0066] Specifically, in this embodiment of the invention, the CTA image of the plaque phantom and the CTA image to be detected are CTA images taken by the same device under the same conditions.

[0067] In one optional implementation, the matching module specifically includes:

[0068] The analysis submodule is used to analyze the CTA images of multiple patch phantoms and the CTA image to be detected using analysis algorithms to obtain the feature vector combination matrix corresponding to the CTA images of multiple patch phantoms and the feature vector combination matrix of the CTA image to be detected.

[0069] The combination submodule is used to combine the feature vector combination matrix corresponding to the CTA image of each patch body model with the feature vector combination matrix of the CTA image to be detected, respectively, to obtain multiple element pairs;

[0070] The evaluation submodule is used to evaluate the similarity of each element pair using an index evaluation function, obtain an evaluation score for each element pair, and select the patch phantom corresponding to the element pair with the highest evaluation score as the target phantom that matches the CTA image to be detected.

[0071] Specifically, the evaluation submodule is used for:

[0072] By using multiple evaluation functions, the similarity of each element pair is evaluated, and the score of each element pair under different evaluation functions is obtained.

[0073] The scores of each element pair under different indicator evaluation functions are weighted and calculated to obtain the comprehensive evaluation score of each element pair;

[0074] The patch phantom corresponding to the element with the highest comprehensive evaluation score is used as the target phantom that matches the CTA image to be detected.

[0075] Specifically, the matching process provided in this embodiment of the invention can be understood in the following way:

[0076] Let the image set of the patch phantom be {T1, T2, ..., T}. m The analysis algorithm can be viewed as a set of functions F = {f1, f2, ..., f}. n The evaluation function is considered as a set of functions G = {g1, g2, ..., g}. n}, the patient's CTA image to be tested is A. Where for each element f of F... i The input is an image (including the image of the patch phantom and the CTA image to be detected), and the output can be a feature vector combination matrix. The elements g of G... i The input is given by f i The output results consist of element pairs (by T) i The corresponding eigenvector combination matrix is ​​obtained by combining the eigenvector combination matrix of A with the eigenvector combination matrix of A. The output is a score of the similarity between the two (e.g., mean squared error, distance under the 2 norm, etc.). Finally, for all scores (for A and phantom T)... i The evaluation indicators include: g1(f1(A), f1(T) i )), g2(f2(A),f2(Ti)), …g n (f n (A),f n (T i The weighted calculation is performed to obtain the comprehensive evaluation score for each element pair, and finally the best comprehensive score T is selected. i The representative phantom is the target phantom that matches A.

[0077] In this embodiment of the invention, each evaluation indicator can be assigned a corresponding weight based on the doctor's experience and statistical experience.

[0078] In this embodiment of the invention, after matching is completed, it can be assumed that the information of the plaque in the patient's CTA image to be detected is similar to the information of the matched target phantom. Therefore, the plaque can be reconstructed using the phantom parameters used when designing the phantom. For example, the true size of the plaque in the target phantom can be determined based on the phantom parameters. The magnification ratio of the plaque on the CTA and the magnified artifact contours can also be determined. Therefore, the plaque can be proportionally reduced in size, and edges with similar contours can be removed as artifacts to reconstruct the true size of the plaque.

[0079] In one optional implementation, the restoration module 104 is specifically used to: obtain image information of the CTA image of the target phantom, restore it by digital image processing means, and obtain the true size of the calcified plaque in the CTA image. The image information includes one or more of the following: phantom point spread function, phantom artifact proportion, and phantom artifact contour.

[0080] Specifically, in this embodiment of the invention, image information such as the phantom point spread function, the proportion of phantom artifacts, and the contour of phantom artifacts can be estimated using the target phantom, and then restored using digital image processing techniques (filtering, inverse transformation, etc.) to obtain the true size of the calcified plaques in the CTA image. Taking the point spread function restoration method as an example, its principle is briefly described as follows:

[0081] Generally, the blurred image g is formed as follows: g(x,y) = f(x,y) × h(x,y) + n(x,y), where f is the true image, h is the point spread function of the device, and n is additive noise. After estimating the device's h using a phantom, ignoring the additive noise n, a CTA with reduced artifacts can be re-detected using patient CT, thus restoring the true size of the plaque. Other methods are similar, directly obtaining image information from the phantom and then using image processing methods to directly restore the image.

[0082] In another optional implementation, the restoration module 104 is specifically used to: determine the true size of the patch in the CTA image to be detected by using a pre-trained restoration model and the CTA image of the target phantom;

[0083] The domain of the restoration algorithm F is the CTA image and the labeled data, and the range is a set G consisting of the initial model and all its parameter combination models. The element g∈G is a function whose domain and range are both CTA image matrices, F(a,b)=g, g(c)=d, where a∈A, b∈B, c is the CTA to be detected, d is the restored CTA, A is the CTA image, and B is the labeled data.

[0084] In this embodiment of the invention, the restoration model is trained according to the following steps:

[0085] S1, manufacture a plaque phantom, and use the phantom information of the plaque phantom as sample phantom annotation data. The sample phantom annotation data includes: phantom form, calcification ratio, plaque size, and plaque type.

[0086] S2, obtain the CTA image of each patch phantom as a sample CTA image;

[0087] S3, the preset model is trained using the sample CTA image and the sample phantom annotation data to obtain the restored model.

[0088] In this embodiment of the invention, a machine learning model can be trained using information such as the actual size of the phantom and phantom images, and the model can be used to reconstruct the actual size of plaques on a patient's CTA.

[0089] For example, for each phantom's CT image, the phantom's true contour can be annotated on the CT image as a label (i.e., the edges of the true contour are drawn with lines). This annotation is used to train a machine learning model, enabling the model to annotate the true contour of patches on general CT images.

[0090] For example, for each phantom's CT image, the phantom's true contour can be annotated on the image as a label. A large number of unlabeled patient CTA images paired with the phantoms are selected (one phantom paired with multiple images). First, a machine learning model is trained using the labeled phantom images. This model is then used to annotate the unlabeled patient images with the true contours of plaques, i.e., pseudo-labels. The pseudo-labels are added to the training set to retrain a new machine learning model. This model is then used to annotate more unlabeled patient images. The above steps are repeated several times, and the model with the best performance is selected.

[0091] For example, for image A to be identified, its paired target phantom can be used to delineate the area of ​​the phantom near the CT scan, treating it as a degraded image. The true contour and size of the phantom are calculated at the same angle and distance, and the PSF function at each location is calculated through inverse Fourier transform, etc., and finally fitted.

[0092] In this embodiment of the invention, when designing the plaque phantom, it can be specifically designed for different target populations. Therefore, the resulting phantom-based coronary CTA calcified plaque detection system will be more targeted towards these populations. For example, if the target population is individuals over 60 years old, whose lesions may generally have a high degree of calcification, the overall range of the calcification ratio parameter can be made higher and narrower, and more precise. For instance, if the original calcification ratio parameter is 10%, 20%, ... 100%, it can be set to 55%, 60%, ... 100% for this population, thereby improving the overall detection efficiency and accuracy of the system. In this case, during algorithm analysis, various patient-specific information (e.g., age, gender, biochemical indicators) can be incorporated as prior knowledge to further improve the overall detection efficiency and accuracy of the system.

[0093] In this embodiment of the invention, a supervised patch-related algorithm can also be set in the system, such as annotating the true contour of the patch to remove artifacts. Specifically, supervised training can be performed using the true contour of the patch phantom and the CTA image of the patch phantom to obtain the corresponding algorithm. This algorithm can then be used to remove artifacts from the CTA image to be detected, further improving the overall detection efficiency and accuracy of the system.

[0094] Reference Figure 2 The figure shows a flowchart of a coronary CTA calcified plaque detection method based on a phantom according to an embodiment of the present invention. As shown in the figure, the coronary CTA calcified plaque detection method based on a phantom provided by the present invention includes the following steps:

[0095] Step 1: Obtain CTA images of multiple patch phantoms;

[0096] Step 2: Acquire the CTA image to be detected;

[0097] Step 3: Using a pre-trained matching model algorithm, match the CTA image of the patch phantom with the CTA image to be detected to obtain the target phantom that matches the CTA image to be detected;

[0098] Step 4: Reconstruct the size of the calcified plaques in the CTA image to be detected based on the target phantom.

[0099] Optionally, step 4 includes:

[0100] The image information of the CTA image of the target phantom is obtained and restored by digital image processing to obtain the true size of the calcified plaques in the CTA image. The image information includes one or more of the following: phantom point spread function, phantom artifact proportion, and phantom artifact contour.

[0101] Optionally, step 4 further includes: determining the true size of the patch in the CTA image to be detected by using a pre-trained reconstruction model and the CTA image of the target phantom;

[0102] The domain of the restoration algorithm F is the CTA image and the labeled data, and the range is a set G consisting of the initial model and all its parameter combination models. The element g∈G is a function whose domain and range are both CTA image matrices, F(a,b)=g, g(c)=d, where a∈A, b∈B, c is the CTA to be detected, d is the restored CTA, A is the CTA image, and B is the labeled data.

[0103] Optionally, the restoration model is trained according to the following steps:

[0104] A plaque phantom is manufactured, and the phantom information of the plaque phantom is used as sample phantom annotation data. The sample phantom annotation data includes: phantom form, calcification ratio, plaque size, and plaque type.

[0105] Obtain CTA images of each patch phantom as sample CTA images;

[0106] The preset model is trained using the sample CTA image and the sample phantom annotation data to obtain the restored model.

[0107] Optionally, step 3 includes:

[0108] The CTA images of multiple patch phantoms and the CTA image to be detected are analyzed using an analysis algorithm to obtain the feature vector combination matrix corresponding to the CTA images of multiple patch phantoms and the feature vector combination matrix of the CTA image to be detected.

[0109] The feature vector combination matrix corresponding to the CTA image of each patch body model is combined with the feature vector combination matrix of the CTA image to be detected to obtain multiple element pairs;

[0110] The similarity of each element pair is evaluated using an index evaluation function, and an evaluation score is obtained for each element pair. The patch phantom corresponding to the element pair with the highest evaluation score is used as the target phantom that matches the CTA image to be detected.

[0111] Optionally, the similarity of each element pair is evaluated using an index evaluation function to obtain an evaluation score for each element pair. The patch phantom corresponding to the element pair with the highest evaluation score is used as the target phantom that matches the CTA image to be detected, including:

[0112] By using multiple evaluation functions, the similarity of each element pair is evaluated, and the score of each element pair under different evaluation functions is obtained.

[0113] The scores of each element pair under different indicator evaluation functions are weighted and calculated to obtain the comprehensive evaluation score of each element pair;

[0114] The patch phantom corresponding to the element with the highest comprehensive evaluation score is used as the target phantom that matches the CTA image to be detected.

[0115] Optionally, the CTA image of the plaque phantom and the CTA image to be detected are CTA images taken by the same device under the same conditions.

[0116] Based on the same inventive concept, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the coronary CTA calcified plaque detection method based on a phantom described in any of the above embodiments.

[0117] Based on the same inventive concept, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the coronary CTA calcified plaque detection method based on a phantom described in any of the above embodiments.

[0118] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0119] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable phantom-based coronary CTA calcification plaque detection terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable phantom-based coronary CTA calcification plaque detection terminal device, generate instructions for implementing the process... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0121] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable phantom-based coronary CTA calcification plaque detection terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0122] These computer program instructions can also be loaded onto a computer or other programmable phantom-based coronary CTA calcified plaque detection terminal device, causing a series of operational steps to be executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal device for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0123] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0124] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0125] The foregoing has provided a detailed description of the coronary CTA calcified plaque detection system, method, equipment, and medium based on a phantom provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A phantom-based coronary CTA calcified plaque detection system, characterized in that, The system includes: The phantom CTA image acquisition module is used to acquire CTA images of multiple patch phantoms; The coronary artery CTA image acquisition module is used to acquire the CTA image to be detected; The matching module is used to match the CTA image of the patch phantom and the CTA image to be detected using a pre-trained matching model algorithm, so as to obtain the target phantom that matches the CTA image to be detected; The restoration module is used to restore the size of the calcified plaques in the CTA image to be detected based on the target phantom; The restoration module is specifically used for: The image information of the CTA image of the target phantom is obtained and restored by digital image processing to obtain the true size of the calcified plaques in the CTA image. The image information includes one or more of the following: phantom point spread function, phantom artifact proportion, and phantom artifact contour.

2. The phantom-based coronary CTA calcified plaque detection system of claim 1, wherein, The restoration module is further configured to: determine the true size of the patch in the CTA image to be detected by using a pre-trained restoration model and the CTA image of the target phantom; The domain of the restoration algorithm F is the CTA image and the labeled data, and the range is a set G consisting of the initial model and all its parameter combinations. The elements g∈G are functions whose domain and range are both CTA image matrices, F(a,b)=g, g(c)=d, where a∈A, b∈B, c is the CTA to be detected, d is the restored CTA, A is the CTA image, and B is the labeled data.

3. The phantom-based coronary CTA calcified plaque detection system of claim 2, wherein, The restoration model was trained according to the following steps: A plaque phantom is manufactured, and the phantom information of the plaque phantom is used as sample phantom annotation data. The sample phantom annotation data includes: phantom form, calcification ratio, plaque size, and plaque type. Obtain CTA images of each patch phantom as sample CTA images; The preset model is trained using the sample CTA image and the sample phantom annotation data to obtain the restored model.

4. The phantom-based coronary CTA calcified plaque detection system of claim 1, wherein, The matching module specifically includes: The analysis submodule is used to analyze the CTA images of multiple patch phantoms and the CTA image to be detected using analysis algorithms to obtain the feature vector combination matrix corresponding to the CTA images of multiple patch phantoms and the feature vector combination matrix of the CTA image to be detected. The combination submodule is used to combine the feature vector combination matrix corresponding to the CTA image of each patch body model with the feature vector combination matrix of the CTA image to be detected, respectively, to obtain multiple element pairs; The evaluation submodule is used to evaluate the similarity of each element pair using an index evaluation function, obtain an evaluation score for each element pair, and select the patch phantom corresponding to the element pair with the highest evaluation score as the target phantom that matches the CTA image to be detected.

5. The coronary CTA calcified plaque detection system based on a phantom according to claim 4, characterized in that, The evaluation submodule is specifically used for: By using multiple evaluation functions, the similarity of each element pair is evaluated, and the score of each element pair under different evaluation functions is obtained. The scores of each element pair under different indicator evaluation functions are weighted and calculated to obtain the comprehensive evaluation score of each element pair; The patch phantom corresponding to the element with the highest comprehensive evaluation score is used as the target phantom that matches the CTA image to be detected.

6. The coronary CTA calcified plaque detection system based on a phantom according to claim 1, characterized in that, The CTA image of the plaque phantom and the CTA image to be detected were taken by the same device under the same conditions.

7. A method for detecting coronary CTA calcified plaques based on a phantom, characterized in that, The method includes: Step 1: Obtain CTA images of multiple patch phantoms; Step 2: Acquire the CTA image to be detected; Step 3: Using a pre-trained matching model algorithm, match the CTA image of the patch phantom with the CTA image to be detected to obtain the target phantom that matches the CTA image to be detected; Step 4: Reconstruct the size of the calcified plaques in the CTA image to be detected based on the target phantom, including: obtaining the image information of the CTA image of the target phantom, and reconstructing it using digital image processing methods to obtain the true size of the calcified plaques in the CTA image. The image information includes one or more of the following: phantom point spread function, phantom artifact proportion, and phantom artifact contour.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the coronary CTA calcified plaque detection method based on a phantom as described in claim 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the coronary CTA calcification plaque detection method based on a phantom as described in claim 7.

Images