Urinary calculus CT image enhancement processing method

Abstract

The invention discloses a urinary calculus CT image enhancement processing method, and relates to the technical field of image processing, and the method comprises the steps: exporting an initial CT image from a DICOM file, and carrying out the sampling of the initial CT image through K sparse angles, and obtaining K sampling CT images; taking K sparse angles as indexes, performing CT artifact image and artifact mapping retrieval, and obtaining a sample CT artifact image-sample artifact group set; constructing an artifact recognizer; performing artifact recognition on the K sampling CT images by using an artifact recognizer, and performing strip artifact separation on the K sampling CT images according to a recognition result to obtain K separated sampling CT images; and carrying out iterative correlation interaction analysis on the K separated and sampled CT images to obtain a target enhanced CT image. The technical problems that artifacts exist in urinary calculus CT images in the prior art, and calculus areas cannot be recognized clearly are solved. The technical effect of enhancing the image quality is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a urinary stone CT image enhancement processing method. Background Art

[0002] Urinary stones are a common disease of the urinary system. CT images, as a diagnostic tool, have been widely used for stone detection and location. However, during low-dose CT scans, sparse sampling angles often lead to image quality degradation and severe streak artifacts, which hinder the accurate diagnosis of stones.

[0003] Existing CT image enhancement methods often suffer from several drawbacks when processing low-dose or sparsely sampled CT images: First, they cannot effectively identify and remove streak artifacts in the image. Second, they provide suboptimal enhancement of urinary stone areas, resulting in loss or blurring of details in the stone area. Especially when the CT image projection angle is small, traditional image enhancement methods often struggle to balance artifact removal with the preservation of the stone area, impacting clinical effectiveness. Therefore, how to effectively remove artifacts and enhance image quality in the stone area has become an urgent problem to be solved. Summary of the Invention

[0004] The present application provides a urinary stone CT image enhancement processing method, which is used to solve the technical problem in the prior art that urinary stone CT images have artifacts and cannot clearly identify the stone area.

[0005] In view of the above problems, the present application provides a method for enhancing CT images of urinary stones, the method comprising: Exporting an initial CT image from a DICOM file, and sampling the initial CT image using K sparse angles to obtain K sampled CT images, where K is a positive integer; Using the K sparse angles as indexes, performing CT artifact image and artifact mapping retrieval to obtain a sample CT artifact image-sample artifact group set; Based on the mapping learning between sample CT artifact images and sample artifact groups, an artifact identifier is constructed. Performing artifact recognition on the K sampled CT images using the artifact recognizer, and performing strip artifact separation on the K sampled CT images according to the recognition result to obtain K separated sampled CT images; Perform iterative correlation interaction analysis on the K separated sampled CT images to obtain a target enhanced CT image.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application exports an initial CT image from a DICOM file and samples the initial CT image using K sparse angles to obtain K sampled CT images, where K is a positive integer. CT artifact images and artifact mappings are then retrieved using the K sparse angles as indexes to obtain a sample CT artifact image-sample artifact group set. Mapping learning is then performed based on the sample CT artifact image-sample artifact group set to construct an artifact identifier. The artifact identifier is then used to perform artifact identification on the K sampled CT images. Strip artifacts are then separated from the K sampled CT images based on the identification results to obtain K separated sampled CT images. Iterative correlation interaction analysis is then performed on the K separated sampled CT images to obtain a target enhanced CT image. This achieves the technical effect of improving the quality of CT image enhancement processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Attachment Figure 1 The figure is a flow chart of a method for enhancing CT images of urinary stones provided by an embodiment of the present invention.

[0008] Attachment Figure 2 It is a flow chart of obtaining a sample CT artifact image-sample artifact group set in a urinary stone CT image enhancement processing method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0009] Below in conjunction with specific embodiment, further set forth the present invention.Should be understood that these embodiments are only used to illustrate the present invention and are not used in limiting the scope of the present invention.In addition, should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall equally within the scope limited by the appended claims of the application.

[0010] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0011] Examples, such as the attached Figure 1 As shown, the present application provides a method for enhancing CT images of urinary stones, wherein the method comprises: S1: exporting an initial CT image from a DICOM file, and sampling the initial CT image using K sparse angles to obtain K sampled CT images, where K is a positive integer; In one possible embodiment, DICOM (Digital Imaging and Communications in Medicine) is a standard format in the field of medical imaging, used to store, transmit, and manage medical imaging data. DICOM files contain image data, patient information, examination details, and other information. The initial CT image refers to the original CT scan image extracted from the DICOM file, which typically includes a cross-sectional image of the patient's urinary system. The sparse angle refers to the selection of a small number of projection angles for image sampling during the CT scanning process. Traditional CT scanning uses more angles to acquire image data, while sparse angles are sampled while reducing the scanning angles to simulate low-dose CT.

[0012] Preferably, the initial CT image is first exported from the DICOM file to ensure the acquisition of the original medical imaging data. The initial CT image is then sampled by selecting K sparse angles. This operation reduces the number of sampled angles while retaining sufficient projection data to ensure the basic integrity of the image information. K is the number of selected angles. This step provides clean, limited, but representative input data for subsequent image enhancement and artifact suppression, simulating low-dose scanning conditions and ensuring that the entire processing flow conforms to the characteristics of low-dose CT images.

[0013] S2: Using the K sparse angles as indexes, perform CT artifact image and artifact mapping retrieval to obtain a sample CT artifact image-sample artifact group set; In one possible embodiment, the CT artifact image refers to a false image or unreal structure that appears in the CT image due to insufficient sampling angles, equipment errors, or algorithm limitations. In low-dose or sparse angle scanning, strip artifacts are the most common type, manifested as regular stripes or unnatural structures in the image. Artifacts are artifact parts that are identified and extracted from the CT artifact image. In this step, a mapping search is performed between CT artifact images and artifacts using K sparse angles as indices to obtain the sample CT artifact image-sample artifact group set. Each sample CT artifact image-sample artifact group includes a sample CT artifact image and a corresponding sample artifact.

[0014] Each sample CT image in the sample CT artifact image-sample artifact group set is paired with its corresponding artifact information (such as the shape and location of the stripe artifact). This set is used to provide reference and training data for subsequent artifact recognition.

[0015] Further, such as Figure 2As shown, the K sparse angles are used as indexes to perform CT artifact image and artifact mapping retrieval to obtain a sample CT artifact image-sample artifact group set. Step S2 of the embodiment of the present application further includes: Obtain K sparse angles as indexes, perform CT artifact image and artifact mapping retrieval in the big data, and obtain an initial sample CT artifact image-sample artifact group set; The sample quality authentication of the initial sample CT artifact image-sample artifact group set is performed from two dimensions: CT artifact image similarity and artifact similarity. If the authentication is passed, the initial sample CT artifact image-sample artifact group set is used as the sample CT artifact image-sample artifact group set.

[0016] Furthermore, step S2 of the embodiment of the present application further includes: Selecting M sample CT artifact images-sample artifact groups from the initial sample CT artifact image-sample artifact group set in a random manner without replacement, where M is a positive integer; Taking M sample CT artifact images-sample artifact groups as M cluster centers, cluster gain analysis is performed on the initial sample CT artifact image-sample artifact group set from two dimensions: CT artifact image similarity and artifact similarity, to obtain the sample clustering gain; Determine whether the sample clustering gain is less than or equal to a preset clustering gain threshold; if so, the authentication is passed.

[0017] Furthermore, with M sample CT artifact images - sample artifact groups as M cluster centers, cluster gain analysis is performed on the initial sample CT artifact image - sample artifact group set from two dimensions: CT artifact image similarity and artifact similarity, to obtain a sample clustering gain. In this embodiment of the application, step S2 further includes: Obtain a sample clustering gain analysis function, wherein the sample clustering gain analysis function is: ; in, is the sample clustering gain, for The first cluster center Cluster sample CT artifact image-sample artifact group set with cluster centers, For the Sample CT artifact images of cluster centers, For the Clustering sample CT artifact image of cluster centers - the first Clustered sample CT artifact images - sample CT artifact images of the sample artifact group, For the The sample artifacts of cluster centers, For the Clustering sample CT artifact image of cluster centers - the first Clustered sample CT artifact images - sample artifacts of the sample artifact group, To balance the weight of CT artifact image similarity and artifact similarity; Clustering the initial sample CT artifact image-sample artifact group set based on M cluster centers to obtain M clustered sample CT artifact image-sample artifact group sets; A sample clustering gain analysis function is used to perform clustering gain analysis on the M cluster centers and the M clustered sample CT artifact image-sample artifact group sets to obtain the sample clustering gain amount.

[0018] In one possible embodiment, K sparse angles are first selected as indices. These angles are capable of simulating image data obtained under low-dose CT scanning conditions. These sparse angles are then used to retrieve corresponding CT images from a large dataset. These images may contain varying degrees of artifacts during their generation. These retrieved images, along with artifact information (e.g., streak artifacts, structural artifacts, etc.), constitute the initial sample CT artifact image set—the sample artifact group. This set provides data support and an analytical foundation for subsequent image enhancement and artifact removal. This mapping retrieval effectively selects artifact-related samples from a large dataset, laying a solid foundation for subsequent artifact identification and removal.

[0019] Furthermore, to obtain a richer sample base and avoid a limited sample coverage that affects subsequent artifact identifier training, the retrieved initial sample CT artifact image-sample artifact group set needs to undergo sample quality certification. If the certification passes, that is, if the initial sample CT artifact image-sample artifact group set contains a sufficiently diverse set of sample types, the initial sample CT artifact image-sample artifact group set can be used as the sample CT artifact image-sample artifact group set, thereby providing data support for subsequent artifact identifier training.

[0020] If the authentication fails, that is, when the initial sample CT artifact image-sample artifact group set does not contain enough sample types, it is necessary to retrieve data samples from the big data again using the K sparse angles as indexes, and re-authenticate the retrieved data samples for sample quality.

[0021] Preferably, when performing retrieval sample quality certification, identification is primarily conducted based on two dimensions: CT artifact image similarity and artifact similarity. Certification can be passed when the sample clustering gain, i.e., the overall similarity of the initial sample CT artifact image and the sample artifact group set, is low. CT artifact image similarity refers to the similarity between the artifact region in a CT image and other known artifact images. This similarity is typically assessed by calculating the distance between image features (e.g., Euclidean distance). The higher the similarity, the more similar the artifact pattern in the current image is to the artifact images in the sample CT artifact image or the sample artifact group at the cluster center. Artifact similarity focuses on the degree of match between artifact features in the image and the sample artifacts in the sample CT artifact image or the sample artifact group at the cluster center. Artifact similarity calculation compares the similarity between the current artifact and the reference artifact by analyzing artifact features such as morphology and intensity in the image, thereby determining whether the artifact types are similar.

[0022] In one possible embodiment, the random sampling without replacement method is a commonly used random sampling method, in which samples will not be replaced again after being selected, so each sample has only one chance to be selected. Specifically, M sample CT artifact images-sample artifact groups are randomly selected from the initial sample CT artifact image-sample artifact group set, and the same sample CT artifact image-sample artifact group will not be selected again after each selection. The M randomly selected sample CT artifact images-sample artifact groups are used as M cluster centers. In the clustering algorithm, the cluster center refers to the center point representing a certain class. In this step, the M selected sample CT artifact images-sample artifact groups are used as the initial centers of the clusters. These centers are used in subsequent cluster analysis to help determine the similarity and distribution between samples. Cluster gain analysis refers to the use of image features (such as CT artifact image similarity and artifact similarity) to evaluate the clustering effect during the clustering process. The sample clustering gain refers to the increase in similarity between clustered samples or the improvement in overall classification performance, measured by analyzing the clustering results of the samples during the clustering process. It reflects the effectiveness of the clustering operation. A higher sample clustering gain indicates that the initial sample CT artifact image-sample artifact group set has too few types and high similarity, which is not conducive to subsequent artifact identifier training.

[0023] Preferably, the preset clustering gain threshold is a standard value pre-set by those skilled in the art to determine whether the sample set has the desired effect. If the sample clustering gain is less than or equal to this threshold, the sample quality is considered to be sufficiently high. If the sample clustering gain is less than the preset clustering gain threshold, it indicates that the sample is well dispersed and can pass the authentication and proceed to the next step of processing.

[0024] Optionally, a sample clustering gain analysis function is used to analyze the overall similarity of the initial sample CT artifact image-sample artifact group set from two dimensions: CT artifact image similarity and artifact similarity. The cosine similarity formula is used to calculate the similarity between each initial sample CT artifact image-sample artifact group and each cluster center in the initial sample CT artifact image-sample artifact group set, and the group is assigned to the cluster containing the cluster center with the greatest similarity, thereby obtaining the M clustered sample CT artifact image-sample artifact group sets. The CT artifact images and their artifact information in each cluster set have high similarity in certain features (such as CT artifact image similarity and artifact similarity).

[0025] The purpose of clustering is to reduce data complexity and improve the efficiency of subsequent analysis by grouping samples. Clustering allows us to identify subsets of samples with similar characteristics from the entire sample set, allowing for more detailed processing and analysis within these subsets. Ultimately, the resulting set of M clustered sample CT artifact images—sample artifact groups—provides high-quality, low-redundancy sample data for subsequent artifact identification and removal, making the process more accurate and efficient.

[0026] Furthermore, the sample clustering gain analysis function is used to quantitatively analyze the overall similarity between the M cluster centers and the M clustered sample CT artifact image-sample artifact group sets to obtain the sample clustering gain.

[0027] S3: Mapping learning is performed based on the sample CT artifact image-sample artifact group set to build an artifact identifier; Furthermore, mapping learning is performed based on the sample CT artifact image-sample artifact group set to construct an artifact identifier. Step S3 of the embodiment of the present application further includes: Each sample artifact in the sample CT artifact image-sample artifact group set is identified, and a convolutional neural network-based framework is trained using the sample CT artifact image-sample artifact group set. The identified sample artifacts are used to supervise the learning process, and the trained artifact identifier is obtained by minimizing the loss function.

[0028] Furthermore, the loss function is: ; in, is the loss amount output by the loss function, is the sample artifact, is the artifact output by the artifact identifier, is the sample artifact group of the i-th sample CT artifact image-sample artifact group in the set of sample CT artifact images-sample artifact groups, is the artifact output by the artifact identifier after identifying the sample CT artifact image of the i-th sample CT artifact image-sample artifact group, N is the number of sample CT artifact images-sample artifact groups in the set of sample CT artifact images-sample artifact groups, and N is an integer less than or equal to 3.

[0029] In one embodiment, the artifact identifier is a trained neural network model specifically designed to identify artifacts in CT images. It can distinguish between artifacts and normal structures in the image, and identified artifacts can be further processed and removed to improve image quality. Convolutional neural networks are a type of deep learning model widely used in image recognition tasks. By extracting image features through convolutional and pooling layers, they can effectively process image data and extract important information from it. A loss function is a function used to measure the difference between the predicted result and the actual label. The goal is to minimize the loss function so that the model's output is closer to the true label. In this context, the loss function measures the error between the artifact identifier's output and the actual artifact identification. The output value (error) of the loss function is minimized by adjusting the model's weights. This goal is typically achieved through optimization algorithms such as gradient descent.

[0030] Preferably, sample artifacts are first extracted from a set of sample CT artifact images—sample artifact groups—and each sample artifact is labeled. This labeling operation provides supervisory information for subsequent training, determining which areas are artifacts and which are normal. Then, these labeled sample CT artifact images—sample artifact groups—are trained based on a convolutional neural network (CNN) framework. Through multiple layers of convolution and pooling operations, the CNN extracts artifact features from the original image and learns how to distinguish artifacts from normal structures. During training, the labeled sample artifacts provide a supervisory signal, helping the network learn the correct artifact recognition method. By minimizing a loss function (e.g., mean squared error), the training process continuously adjusts the parameters of the CNN network to improve the recognition accuracy of the artifact identifier. The loss function measures the difference between the predicted artifacts and the actual labels. Minimizing this loss means that the network is better able to recognize and process artifacts in the image.

[0031] Ultimately, the trained artifact detector will be able to accurately identify artifact areas in the image and provide support for subsequent image restoration and enhancement. The core role of this process is to build an effective artifact recognition model that can greatly improve the effectiveness and accuracy of image artifact removal.

[0032] S4: performing artifact recognition on the K sampled CT images using the artifact recognizer, and performing strip artifact separation on the K sampled CT images according to the recognition result to obtain K separated sampled CT images; In one embodiment, the K separated sampled CT images are images obtained by separating the stripe-like artifacts from the K sampled CT images. The separated images, with artifacts removed, provide clearer CT image content. First, artifact identification is performed on the K sampled CT images using a trained artifact identifier. Using the artifact identifier, the system can automatically detect areas of artifacts that may exist in each image. The artifact identifier identifies the locations of artifact areas for each CT image; these areas are typically labeled as artifact labels to facilitate subsequent processing. After identification, the stripe-like artifacts in each sampled CT image are separated based on the artifact identification results. Stripe-like artifacts typically appear as lines or strips in CT images and may affect the physician's interpretation of the image. To improve image quality, the system uses image processing techniques (such as filtering, denoising, and image restoration) to separate these artifacts from the image.

[0033] Ultimately, K separated sampled CT images are output—those with artifacts removed or separated. These images are closer to the true lesion image and reduce artifact interference, providing clearer image data for further image analysis and diagnosis. The core purpose of this process is to significantly improve CT image quality through artifact identification and separation, enabling more accurate subsequent image enhancement or analysis.

[0034] S5: Perform iterative correlation interaction analysis on the K separated sampled CT images to obtain a target enhanced CT image.

[0035] Furthermore, iterative correlation interaction analysis is performed on the K separated sampled CT images to obtain a target enhanced CT image. In this embodiment of the application, step S5 further includes: Traversing the K separated sampling CT images to perform feature extraction and obtain K separated sampling CT image feature sets; randomly extracting a first separated sampling CT image feature set and a second separated sampling CT image feature set from the K separated sampling CT image feature sets without replacement; Performing iterative correlation interaction analysis on the first separated sampling CT image feature set and the second separated sampling CT image feature set to obtain a first iterative correlation interaction separated sampling CT image feature set; Performing iterative correlation interaction analysis on the first iterative correlation interactive separated sampling CT image feature set and a third separated sampling CT image feature set randomly extracted without replacement from the K separated sampling CT image feature sets to obtain a second iterative correlation interactive separated sampling CT image feature set; After multiple iterative correlation interaction analyses, a K-1th iterative correlation interaction separation sampling CT image feature set is obtained, which is analyzed with the Kth separation sampling CT image using a forward feedback network to obtain the target enhanced CT image.

[0036] Furthermore, iterative correlation interaction analysis is performed on the first separated sampled CT image feature set and the second separated sampled CT image feature set to obtain a first iterative correlation interaction separated sampled CT image feature set. In this embodiment of the application, step S5 further includes: Performing inner product mapping analysis on the first separated sampling CT image feature set and the second separated sampling CT image feature set to obtain a first iterative correlation interaction feature similarity set; Normalizing the first iterative association interaction feature similarity set, and filling the normalized result into an initially empty matrix to obtain a first iterative association interaction matrix; A convolution operation is performed on the first iterative correlation interaction matrix and the first iterative correlation interaction feature similarity set to obtain the first iterative correlation interaction separation sampling CT image feature set.

[0037] In one possible embodiment, since the K separately sampled CT images come from K different sparse angles, the images reflect different scales of information. The smaller the sparse angle, the larger the scale of information contained in the image, the coarser the information, but it can reflect more contours. The larger the sparse angle, the higher the image detail, but the amount of information contained is too redundant. Therefore, by performing iterative correlation interaction analysis on the K separately sampled CT images, the images sampled at different sparse angles are fused. The images with small sparse angles capture more large-scale structures, while the images with large sparse angles capture more details, thereby fusing information of different scales and achieving the technical effect of effectively improving image detail and quality.

[0038] First, the first separated sampling CT image feature set and the second separated sampling CT image feature set are randomly extracted without replacement from the K separated sampling CT image feature sets. These two feature sets represent the feature data in the two groups of CT images respectively. This operation provides two sets of benchmark data for the subsequent feature association analysis. Next, the two sets of feature sets are subjected to iterative association interaction analysis. Through an iterative approach, the feature sets are continuously optimized and adjusted so that the image features can gradually influence each other in the two sets of data, achieving a higher feature matching degree. The purpose of this process is to improve the similarity and accuracy of image features, especially in CT images of urinary stones, so that the detailed features of the stones can be effectively extracted.

[0039] During the iterative analysis process, inner product mapping analysis is also performed to measure the similarity between feature sets. For example, this analysis can be performed using the cosine similarity formula. This inner product operation can quantitatively assess the similarity between two feature sets, providing a basis for feature optimization. The similarity results are then normalized to ensure that differences between different features are appropriately controlled to avoid bias in subsequent analysis.

[0040] Furthermore, by filling the normalized similarity results into a matrix and performing a convolution operation, the effect of feature extraction and analysis is further enhanced. The convolution operation not only helps with feature extraction, but also creates more detailed associations between different image features, thus providing more accurate input data for the final image enhancement.

[0041] Finally, through multiple iterations of correlation and interaction analysis, image features are repeatedly processed in multiple steps. Each iteration updates the features based on the results of the previous round, gradually optimizing the extraction of image features. This analysis method continuously improves the accuracy and quality of the target image through continuous adjustment. After obtaining the feature set of the K-1th iterative correlation and interaction separated sampling CT image, it is combined with the Kth separated sampling CT image through a forward feedback network analysis to generate the target enhanced CT image. This process continuously optimizes image features through a feedback mechanism, ultimately resulting in a target enhanced CT image that has been enhanced, artifact-free, and optimized for detail.

[0042] Preferably, by obtaining multiple sample iteratively associated interactively separated sampling CT image feature sets and multiple sample separated sampling CT images, as well as multiple corresponding sample enhanced CT images as training data, supervised training is performed on the framework constructed based on the forward neural network until the training converges, thereby obtaining the trained forward feedback network. Furthermore, the trained forward feedback network is used to analyze the K-1th iteratively associated interactively separated sampling CT image feature set and the Kth separated sampling CT image to obtain the target enhanced CT image. CT images sampled from different sparse angles are obtained, and image quality is optimized through iterative analysis and deep learning, ultimately obtaining high-quality target enhanced CT images, thereby improving the detection accuracy and image quality of urinary stones.

[0043] In summary, the embodiments of the present application have at least the following technical effects: 1. This application simulates low-dose CT scans by sampling the initial CT image at K sparse angles, avoiding the image artifacts caused by excessive doses in traditional methods. Furthermore, the artifact detector effectively identifies and removes streak artifacts in the image, improving the clarity and accuracy of the final image.

[0044] 2. By obtaining a set of sample CT artifact images and sample artifact groups from CT images and artifact mappings, and performing mapping learning based on this set, an artifact identifier is constructed, thereby improving the accuracy of artifact recognition. Through iterative correlation interaction analysis, image features can be continuously optimized in each round of processing, making the final target enhanced CT image more accurate and clear.

[0045] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0046] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0047] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A method for enhancing CT images of urinary stones, characterized in that: The method comprises: Exporting an initial CT image from a DICOM file, and sampling the initial CT image using K sparse angles to obtain K sampled CT images, where K is a positive integer; Using the K sparse angles as indexes, performing CT artifact image and artifact mapping retrieval to obtain a sample CT artifact image-sample artifact group set; Based on the mapping learning between sample CT artifact images and sample artifact groups, an artifact identifier is constructed. Performing artifact recognition on the K sampled CT images using the artifact recognizer, and performing strip artifact separation on the K sampled CT images according to the recognition result to obtain K separated sampled CT images; Perform iterative correlation interaction analysis on the K separated sampled CT images to obtain a target enhanced CT image.

2. The method for enhancing CT images of urinary stones according to claim 1, wherein: Performing iterative correlation interaction analysis on the K separated sampled CT images to obtain a target enhanced CT image, including: Traversing the K separated sampling CT images to perform feature extraction and obtain K separated sampling CT image feature sets; randomly extracting a first separated sampling CT image feature set and a second separated sampling CT image feature set from the K separated sampling CT image feature sets without replacement; Performing iterative correlation interaction analysis on the first separated sampling CT image feature set and the second separated sampling CT image feature set to obtain a first iterative correlation interaction separated sampling CT image feature set; Performing iterative correlation interaction analysis on the first iterative correlation interactive separated sampling CT image feature set and a third separated sampling CT image feature set randomly extracted without replacement from the K separated sampling CT image feature sets to obtain a second iterative correlation interactive separated sampling CT image feature set; After multiple iterative correlation interaction analyses, a K-1th iterative correlation interaction separation sampling CT image feature set is obtained, which is analyzed with the Kth separation sampling CT image using a forward feedback network to obtain the target enhanced CT image.

3. The method for enhancing CT images of urinary stones according to claim 2, wherein: Performing iterative correlation interaction analysis on the first separated sampling CT image feature set and the second separated sampling CT image feature set to obtain a first iterative correlation interaction separated sampling CT image feature set, including: Performing inner product mapping analysis on the first separated sampling CT image feature set and the second separated sampling CT image feature set to obtain a first iterative correlation interaction feature similarity set; Normalizing the first iterative association interaction feature similarity set, and filling the normalized result into an initially empty matrix to obtain a first iterative association interaction matrix; A convolution operation is performed on the first iterative correlation interaction matrix and the first iterative correlation interaction feature similarity set to obtain the first iterative correlation interaction separation sampling CT image feature set.

4. The method for enhancing CT images of urinary stones according to claim 1, wherein: Using the K sparse angles as indexes, CT artifact images and artifact mappings are retrieved to obtain a sample CT artifact image-sample artifact group set, including: Obtain K sparse angles as indexes, perform CT artifact image and artifact mapping retrieval in the big data, and obtain an initial sample CT artifact image-sample artifact group set; The sample quality authentication of the initial sample CT artifact image-sample artifact group set is performed from two dimensions: CT artifact image similarity and artifact similarity. If the authentication is passed, the initial sample CT artifact image-sample artifact group set is used as the sample CT artifact image-sample artifact group set.

5. The method for enhancing CT images of urinary stones according to claim 4, wherein: include: Selecting M sample CT artifact images-sample artifact groups from the initial sample CT artifact image-sample artifact group set in a random manner without replacement, where M is a positive integer; Taking M sample CT artifact images-sample artifact groups as M cluster centers, cluster gain analysis is performed on the initial sample CT artifact image-sample artifact group set from two dimensions: CT artifact image similarity and artifact similarity, to obtain the sample clustering gain; Determine whether the sample clustering gain is less than or equal to a preset clustering gain threshold; if so, the authentication is passed.

6. The method for enhancing CT images of urinary stones according to claim 5, characterized in that: Taking M sample CT artifact images-sample artifact groups as M cluster centers, cluster gain analysis is performed on the initial sample CT artifact image-sample artifact group set from two dimensions: CT artifact image similarity and artifact similarity. The sample clustering gain is obtained, including: Obtain a sample clustering gain analysis function, wherein the sample clustering gain analysis function is: ； in, is the sample clustering gain, for The first cluster center Cluster sample CT artifact image-sample artifact group set with cluster centers, For the The sample CT artifact images of cluster centers, For the Clustering sample CT artifact image of cluster centers - the first Clustered sample CT artifact images - sample CT artifact images of the sample artifact group, For the The sample artifacts of cluster centers, For the Clustering sample CT artifact image of cluster centers - the first Clustered sample CT artifact images - sample artifacts of the sample artifact group, To balance the weight of CT artifact image similarity and artifact similarity; Clustering the initial sample CT artifact image-sample artifact group set based on M cluster centers to obtain M clustered sample CT artifact image-sample artifact group sets; A sample clustering gain analysis function is used to perform clustering gain analysis on the M cluster centers and the M clustered sample CT artifact image-sample artifact group sets to obtain the sample clustering gain amount.

7. The method for enhancing CT images of urinary stones according to claim 1, wherein: Mapping learning is performed based on the sample CT artifact image-sample artifact group set to build an artifact identifier, including: Each sample artifact in the sample CT artifact image-sample artifact group set is identified, and a convolutional neural network-based framework is trained using the sample CT artifact image-sample artifact group set. The identified sample artifacts are used to supervise the learning process, and the trained artifact identifier is obtained by minimizing the loss function.

8. The method for enhancing CT images of urinary stones according to claim 7, wherein: The loss function is: ； in, is the loss amount output by the loss function, is the sample artifact, is the artifact output by the artifact identifier, is the sample artifact group of the i-th sample CT artifact image-sample artifact group in the set of sample CT artifact images-sample artifact groups, is the artifact output by the artifact identifier after identifying the sample CT artifact image of the i-th sample CT artifact image-sample artifact group, N is the number of sample CT artifact images-sample artifact groups in the set of sample CT artifact images-sample artifact groups, and N is an integer less than or equal to 3.

Images