Wrist joint ligament image processing and analysis method, device and equipment and storage medium

Abstract

The present application relates to the technical field of three-dimensional medical images, and provides a wrist joint ligament image processing and analysis method, device, equipment and storage medium, comprising: acquiring original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged; performing isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image; performing segmentation on the target MRI image to obtain a target segmentation image labeled with a wrist bone image region and a triangular fibrocartilage complex (TFCC) image region; performing edge extraction on the target segmentation image to obtain a TFCC stop point range on the wrist bone, and acquiring a target quantitative index corresponding to the stop point range. The present application obtains a three-dimensional high-resolution image with coherent structure, real texture and high reliability through isotropic super-resolution reconstruction. In addition, image processing, edge extraction and other methods are comprehensively used to efficiently and accurately extract the stop point range.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional medical image technology, and in particular to a wrist joint ligament image processing and analysis method, device, equipment and storage medium. Background Art

[0002] Triangular Fibrocartilage Complex (TFCC) injuries are the most common cause of wrist pain and instability, often requiring surgical treatment. However, as many as 40% of cases currently have poor outcomes, making this a recognized problem in the international academic community. A key reason for this challenge is that the detailed three-dimensional anatomy of the TFCC ligaments and their complex kinematic and dynamic coordination mechanisms remain largely undetermined, providing a reliable basis for accurate diagnosis and planning anatomically and biomechanically sound treatments. The delicate structure and complex function of wrist ligaments present significant challenges for researchers seeking a comprehensive understanding of their morphological and biomechanical properties.

[0003] Currently, MRI sequences at 1.5T or 3.0T with a spacing of 2 mm or more are often recommended for diagnosis and adjunctive treatment of suspected TFCC injuries. However, these imaging techniques present inherent challenges for accurate clinical diagnosis and subsequent appropriate treatment selection due to thick MRI slices, fewer visible ligament layers, and low resolution, making ligament identification difficult. Summary of the Invention

[0004] The present invention provides a wrist joint ligament image processing and analysis method, device, equipment and storage medium, which are used to solve the defects of the wrist joint ligament image processing and analysis method in the prior art.

[0005] The present invention provides a wrist joint ligament image processing and analysis method, the method comprising:

[0006] Acquiring original magnetic resonance imaging (MRI) images of at least two different views of the wrist joint to be imaged;

[0007] Performing isotropic super-resolution reconstruction on original MRI images of at least two different views to obtain a target MRI image;

[0008] Segmenting the target MRI image to obtain a target segmented image with a carpal bone image region and a triangular fibrocartilage complex (TFCC) image region marked;

[0009] Edge extraction is performed on the target segmented image to obtain the end point range of the TFCC on the carpal bone, and the target quantitative index corresponding to the end point range is obtained.

[0010] According to a wrist ligament image processing and analysis method provided by the present invention, segmenting the target MRI image to obtain a target segmented image with annotated carpal bone image areas and triangular fibrocartilage complex (TFCC) image areas includes:

[0011] The target MRI image is imported into 3D Slicer software, and the universal segmentation model SAM is called in the 3D Slicer software to interactively segment the target MRI image to obtain a target segmentation image with the carpal bone image area and the triangular fibrocartilage complex TFCC image area marked.

[0012] According to a wrist joint ligament image processing and analysis method provided by the present invention, the wrist joint to be imaged includes a plurality of target segmented images in the Z-axis direction. Before edge extraction is performed on the target segmented images, the method further includes:

[0013] Interpolation is performed on the plurality of target segmented images at alternate frames to supplement spatial information of the plurality of target segmented images in the Z-axis direction.

[0014] A wrist joint ligament image processing and analysis method provided by the present invention further includes:

[0015] Based on a joint smoothing algorithm, each target segmented image after interpolation of alternate frames is smoothed.

[0016] According to a wrist joint ligament image processing and analysis method provided by the present invention, edge extraction is performed on the target segmented image to obtain the insertion point range of the TFCC on the wrist bone, including:

[0017] Segmenting the carpal bone image region and the TFCC image region from the target segmented image;

[0018] Acquire a first edge contour of the carpal bone image region and a second edge contour of the TFCC image region;

[0019] Based on the first edge contour and the second edge contour, the intersection area of ​​the carpal bone image area and the TFCC image area is determined to obtain the end point range of the TFCC on the carpal bone.

[0020] According to a wrist joint ligament image processing and analysis method provided by the present invention, obtaining a first edge contour of the carpal bone image region and a second edge contour of the TFCC image region includes:

[0021] Processing the carpal bone image region based on a Sobel operator edge detection algorithm and a morphological dilation and erosion algorithm to obtain a first edge contour of the carpal bone image region; and

[0022] The TFCC image region is processed based on a Sobel operator edge detection algorithm and a morphological expansion and corrosion algorithm to obtain a second edge contour of the TFCC image region.

[0023] According to the wrist ligament image processing and analysis method provided by the application, the target quantitative index corresponding to the stop point range is obtained, and the method comprises the following steps:

[0024] The stop point range is fitted based on a RANSAC algorithm.

[0025] The target quantitative index corresponding to the stop point range is obtained by performing data statistics on the stop point range after plane fitting.

[0026] The application further provides a wrist ligament image processing and analysis device, and the device comprises:

[0027] An image acquisition module is configured to acquire original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged.

[0028] An image reconstruction module is configured to perform isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image.

[0029] An image segmentation module is configured to segment the target MRI image to obtain a target segmentation image in which wrist bone image regions and triangular fibrocartilage complex (TFCC) image regions are labeled.

[0030] A stop point analysis module is configured to perform edge extraction on the target segmentation image to obtain a stop point range of the TFCC on the wrist bone and to obtain a target quantitative index corresponding to the stop point range.

[0031] The application further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the wrist ligament image processing and analysis method according to any one of the above embodiments when executing the program.

[0032] The application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is executable on a processor to implement the wrist ligament image processing and analysis method according to any one of the above embodiments.

[0033] The application further provides a computer program product comprising a computer program, wherein the computer program is executable on a processor to implement the wrist ligament image processing and analysis method according to any one of the above embodiments.

[0034] The wrist ligament image processing and analysis method, device, equipment and storage medium provided by the application, the method comprises the following steps: acquiring original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged; performing isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image; performing segmentation on the target MRI image to obtain a target segmentation image labeled with a wrist bone image region and a triangular fibrocartilage complex (TFCC) image region; performing edge extraction on the target segmentation image to obtain a TFCC stop point range on the wrist bone, and acquiring a target quantitative index corresponding to the stop point range. The application can generate a three-dimensional high-resolution target MRI image with coherent structure, real texture and high credibility by fusing the real multi-view original MRI image through isotropic super-resolution reconstruction, and can solve the problems of large MRI layer thickness of the existing clinical wrist joint, few TFCC layers, low resolution and difficult-to-identify ligament structure, and has important significance for observation and pathological diagnosis of the wrist joint and TFCC fine structure. In addition, the application can efficiently and accurately extract the stop point range and the corresponding target quantitative index by comprehensively using image processing, edge extraction and plane fitting methods, can accurately depict the complex and fine three-dimensional anatomical structure of the TFCC, realize automatic and accurate correspondence between the fine three-dimensional anatomy and the imageology of the TFCC, and guide the diagnosis and treatment of TFCC injury, thereby improving the clinical curative effect. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the 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 are some embodiments of the application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0036] Figure 1 is a flowchart of the wrist ligament image processing and analysis method provided by the application;

[0037] Figure 2 is a structural schematic diagram of the wrist ligament image processing and analysis device provided by the application;

[0038] Figure 3 is a structural schematic diagram of the electronic device provided by the application. DETAILED DESCRIPTION

[0039] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0040] The scientific research methods for the morphology and biomechanics of TFCC at present are still limited to in vitro research of gross anatomy specimens, and there are significant defects such as large measurement error, inaccuracy of correspondence between in vitro dissection data and medical images, difficulty in guiding precise diagnosis and treatment, and inability to reflect the real in vivo biomechanical conditions, which are specifically manifested as follows:

[0041] (1) Under the research condition based on gross anatomy specimens, the measurement of ligament morphology data by the existing traditional method inevitably has errors, and this defect is particularly prominent in the measurement of fine wrist ligaments.

[0042] (2) The most important ligament insertion area mapping data cannot be accurately corresponded to the imaging system, and has limited guiding significance for planning the mechanical center and repair direction of ligament repair.

[0043] (3) In vitro biomechanical research needs to exclude other ligaments, joint capsules, forearm interosseous membranes, joint surrounding muscles, tendons and other tissues outside the target ligament. Although this method is a standard scheme in the research of large joints such as shoulder, knee and hip, the wrist joint is composed of many joints and is not a single joint, and all ligaments, joint capsules and muscle tissues play a greater role in cooperation.

[0044] Therefore, the current in vitro research based on gross anatomy specimens not only has poor reliability, but also is difficult to correspond with imaging data and guide clinical diagnosis and treatment research. Accurately depicting the complex three-dimensional anatomical structure of TFCC and its corresponding function is still an unsolved problem.

[0045] In view of the above problems, the present application provides a wrist ligament image processing and analysis method. Figure 1 The flowchart of the wrist ligament image processing and analysis method provided by the present application is shown in Figure 1 The method comprises the following steps:

[0046] Step 110, acquiring original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged;

[0047] In this embodiment, anisotropic three-dimensional images are acquired under at least two different views (or perspectives), anisotropy meaning that the resolution of scanning in the x, y, and z coordinate directions is not the same. The number of views can be any positive integer, and is typically 1-5. The scanning view direction is preferably a three-dimensional orthogonal view, but other view directions can also be used.

[0048] For example, the original MRI images of the wrist joint to be imaged are collected in the coronal, axial, and sagittal views. The voxel sizes of these original MRI images in different directions are (1, 0.1, 0.1) mm, (0.1, 1, 0.1) mm, and (0.1, 0.1, 1) mm, respectively, i.e., the pixel spacing in different directions is different, resulting in anisotropic image pixels.

[0049] In step 120, isotropic super-resolution reconstruction is performed on the original MRI images of at least two different views to obtain a target MRI image.

[0050] The target MRI image refers to a high-resolution image obtained after processing, and the pixel spacing in each direction is equal, i.e., the MRI image has the characteristic of isotropy.

[0051] In this embodiment, a higher resolution, isotropic target MRI image is generated by combining multiple low-resolution original MRI images in different views using image information fusion. The target MRI image can be reconstructed from the image information in the direction with the highest resolution in each of the multiple different views, so that the reconstructed isotropic three-dimensional image has high resolution, fewer artifacts, and better authenticity.

[0052] In one example, the original MRI images in the coronal, axial, and sagittal views are collected as input, and the voxel sizes of these original MRI images in different directions are (1, 0.1, 0.1) mm, (0.1, 1, 0.1) mm, and (0.1, 0.1, 1) mm, respectively. The original MRI images in different views are registered to ensure that they are spatially aligned. A deep learning model such as a convolutional neural network is used to extract features from the registered MRI images, and the extracted features are fused together from the MRI images in different views to obtain a more comprehensive and rich information representation. An isotropic super-resolution algorithm is used to reconstruct the fused MRI images to generate an isotropic high-resolution target MRI image with the same spatial resolution in three directions, and the size of each voxel is 0.1 mm.

[0053] Step 130, segmenting the target MRI image to obtain a target segmentation image labeled with the image region of the carpal bone and the image region of the triangular fibrocartilage complex (TFCC);

[0054] In this embodiment, the segmentation model trained by fine-tuning using the wrist TFCC dataset can be used to segment the target MRI image to obtain the carpal bone and TFCC region, such as U-Net, Mask R-CNN, etc. The segmentation model will automatically identify and label the carpal bone and TFCC region in the target MRI image.

[0055] Step 140, edge extraction is performed on the target segmentation image to obtain the TFCC stop point range on the carpal bone, and the target quantitative index corresponding to the stop point range is obtained.

[0056] The target quantitative index includes but is not limited to morphological indicators (such as length, width, height, ellipsoid degree) and the like, and is not limited thereto.

[0057] Specifically, the edge detection algorithm such as Canny edge detection, Sobel operator, etc. can be used to perform edge detection on the target segmentation image to obtain the carpal bone edge and the TFCC edge, so as to obtain the TFCC stop point range on the carpal bone.

[0058] After obtaining the TFCC stop point range on the carpal bone, the target quantitative index can be calculated for the region, for example, the TFCC stop point range volume, surface area and other quantitative indicators are calculated to describe the distribution range of the TFCC on the carpal bone. Based on the obtained quantitative indicators, further analysis and comparison can be performed.

[0059] The method provided by the embodiment of the present application comprises the following steps: acquiring original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged; performing isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image; performing segmentation on the target MRI image to obtain a target segmentation image in which wrist bone image regions and triangular fibrocartilage complex (TFCC) image regions are labeled; performing edge extraction on the target segmentation image to obtain a TFCC enthesis range on the wrist bone, and acquiring a target quantitative index corresponding to the enthesis range. The present application can generate a three-dimensional high-resolution target MRI image with coherent structure, real texture and high credibility by fusing the original MRI images of multiple views through isotropic super-resolution reconstruction, and can solve the problems of large layer thickness of the existing clinical wrist joint MRI, fewer layers of the visible TFCC, low resolution and difficult-to-identify ligament structure, and has important significance for observation and pathological diagnosis of fine structures such as wrist joints and TFCC. In addition, the present application can accurately depict the complex and fine three-dimensional anatomical structure of TFCC and realize automatic and accurate correspondence between fine three-dimensional anatomy and imaging of TFCC by efficiently and accurately extracting the enthesis range and the corresponding target quantitative index through comprehensive use of image processing, edge extraction, plane fitting and other methods, so as to guide the diagnosis and treatment of TFCC injury and improve the clinical efficacy.

[0060] According to any one of the above embodiments, the segmentation of the target MRI image to obtain a target segmentation image in which wrist bone image regions and triangular fibrocartilage complex (TFCC) image regions are labeled comprises:

[0061] The target MRI image is imported into 3D Slicer software, and a general segmentation model (SAM) is called in the 3D Slicer software to perform interactive segmentation on the target MRI image to obtain a target segmentation image in which wrist bone image regions and triangular fibrocartilage complex (TFCC) image regions are labeled.

[0062] 3D Slicer is an open-source medical image processing software mainly used for visualization, segmentation, reconstruction and analysis of medical image data. It supports multiple image formats, including DICOM, NIFTI, MHA, etc., and provides multiple segmentation algorithms and tools that can be used for segmentation and reconstruction of different anatomical structures.

[0063] The general segmentation model (Segment Anything Model, SAM) is a general deep learning model designed to achieve image segmentation tasks of various different types of targets. The SAM model adopts advanced deep learning architectures such as U-Net, Mask R-CNN, etc., has strong generalization ability and applicability, and can be applied to target segmentation tasks in different fields and application scenarios.

[0064] In this embodiment, the running environment of the SAM model is installed and configured before use, such as the deep learning framework, related libraries and dependencies required by the SAM model. Prepare the wrist TFCC dataset, including labeled data and images. Fine-tune the SAM model using the TFCC dataset to improve the accuracy and generalization ability of the SAM model for the TFCC segmentation task. Install and configure the Samm Base plugin, connect the fine-tuned SAM model to provide segmentation services. Configure plugin parameters and interfaces to ensure normal communication with the SAM model.

[0065] Next, the target MRI image is imported into the 3D Slicer software, and the fine-tuned SAM model is used to provide segmentation services to realize interactive segmentation of TFCC and wrist joints in the 3D Slicer software.

[0066] Specifically: first, configure the Samm Base plugin to provide segmentation services using the fine-tuned SAM model; after importing the target image into the 3D Slicer software, select the target frame, and click the Compute Embedding button in the plugin UI to calculate the image encoding, that is, the image is transmitted to the SAM model for encoding; then, based on the UI page of Slicer, interactive segmentation is performed: after completing the encoding, click the Start Mask Sync button in the Prompts section on the left side of the 3D Slicer software to start the interaction. The green dot function in the Samm Base plugin is "select target area", and the red dot function is "remove area"; the arrow key is to add a point, and the trash can key is to delete a point. The green dot selects the target area of interest, and the red dot removes the redundant area to obtain the segmentation result.

[0067] The method provided by the embodiment of the application uses an interactive segmentation algorithm based on the SAM model to improve the segmentation efficiency while bringing higher segmentation accuracy and more reliable segmentation results. Compared with full-automatic segmentation, interactive segmentation can integrate the user's prior knowledge and integrate the user's judgment information about the existing image into the image segmentation work, thereby improving the segmentation quality. At the same time, the segmentation result can be adjusted through interactive prompt points, making the segmentation process more transparent and reliable.

[0068] Based on any of the above embodiments, the wrist joint to be imaged includes a plurality of target segmentation images in the Z-axis direction, and before the target segmentation images are edge extracted, the method further includes:

[0069] Frame interpolation is performed on the plurality of target segmentation images to supplement the spatial information of the plurality of target segmentation images in the Z-axis direction.

[0070] It should be noted that the target segmentation image is usually the segmentation result of the wrist joint to be imaged on a two-dimensional plane in the Z-axis direction. In order to increase the spatial information of the Z-axis (vertical direction), this embodiment performs interpolation between frames to fill in the missing information of these images in the Z-axis direction, thereby improving the overall spatial information integrity and continuity. Through interpolation between frames, a new target segmentation image can be generated with a higher spatial resolution in the Z-axis direction, so that subsequent processing steps such as edge extraction can be performed more accurately.

[0071] The method provided by the embodiment of the present invention supplements the spatial information of the segmentation result in the Z-axis direction, making the segmentation result more three-dimensional and accurate, thereby better supporting the analysis and application of wrist joint images.

[0072] Based on any of the above embodiments, the further comprising:

[0073] Based on a joint smoothing algorithm, each target segmented image after interpolation of alternate frames is smoothed.

[0074] In this embodiment, after obtaining a segmentation result with added Z-axis spatial information, a joint smoothing algorithm is used to smooth the image. The joint smoothing algorithm is a smoothing method that combines spatial information and pixel value information. It aims to preserve the intersection of the carpal bones and the TFCC while removing jagged edges from the segmentation result, thereby improving image quality. Joint smoothing algorithms include, but are not limited to, bilateral filtering and non-local means filtering, and are not limited to these methods.

[0075] The method provided by the embodiment of the present invention can further optimize and improve the segmentation result by combining the smoothing algorithm, making it more accurate and natural.

[0076] Based on any of the above embodiments, performing edge extraction on the target segmented image to obtain the insertion point range of the TFCC on the carpal bone includes:

[0077] Segmenting the carpal bone image region and the TFCC image region from the target segmented image;

[0078] Acquire a first edge contour of the carpal bone image region and a second edge contour of the TFCC image region;

[0079] Based on the first edge contour and the second edge contour, the intersection area of ​​the carpal bone image area and the TFCC image area is determined to obtain the end point range of the TFCC on the carpal bone.

[0080] In this embodiment, an image segmentation algorithm (such as a semantic segmentation algorithm based on deep learning) can be used to segment the carpal bone image area and the TFCC image area. For the carpal bone image area and the TFCC image area, an edge detection algorithm (such as Canny edge detection) can be used to obtain their edge contours. By comparing and analyzing the edge contours, their intersection area can be determined, thereby obtaining the end point range of the TFCC on the carpal bone.

[0081] Based on any of the foregoing embodiments, obtaining the first edge contour of the carpal bone image area and the second edge contour of the TFCC image area includes:

[0082] Processing the carpal bone image region based on a Sobel operator edge detection algorithm and a morphological dilation and erosion algorithm to obtain a first edge contour of the carpal bone image region; and

[0083] The TFCC image area is processed based on the Sobel operator edge detection algorithm and the morphological dilation and erosion algorithm to obtain a second edge contour of the TFCC image area.

[0084] In this embodiment, the wrist bone image area and the TFCC image area are first converted into grayscale images. Then, the Sobel operator is applied to perform edge detection on the grayscale image to obtain the intensity gradient information of the edge in the image. Based on the edge gradient information obtained by the Sobel operator, an appropriate threshold can be set to filter out the edge area of ​​interest. Morphological dilation and erosion operations are performed on the image after threshold processing to fill the breaks between edges, connect the gaps between edges, and eliminate noise. Finally, based on the image after morphological processing, the first edge contour and the second edge contour can be extracted.

[0085] The method provided in the embodiment of the present invention, by combining the Sobel operator edge detection algorithm and the morphological dilation and erosion algorithm, can effectively extract the edges of the carpal bones and TFCC in the image, thereby improving the quality of medical images and the acquisition of information.

[0086] Based on any of the foregoing embodiments, obtaining a target quantitative indicator corresponding to the stop point range includes:

[0087] Performing plane fitting on the stop point range based on the RANSAC algorithm;

[0088] Data statistics are performed on the dead point range after plane fitting to obtain a target quantitative index corresponding to the dead point range.

[0089] The RANSAC (Random Sample Consensus) algorithm is an iterative method for fitting models and identifying outliers in data.

[0090] In this embodiment, the RANSAC algorithm is used to fit the plane of the TFCC on the carpal bone. The RANSAC algorithm can effectively fit the plane model that best fits the data distribution, so that the data has outliers or noise, so as to improve the accuracy. Then based on the fitted stop point range, data statistics is carried out, so as to obtain the volume, surface area and other quantitative indexes of the stop point range.

[0091] Based on any one of the above embodiments, the present application further provides a wrist ligament image processing and analysis device, Figure 2 is a structural schematic diagram of the wrist ligament image processing and analysis device provided by the present application, as Figure 2 shown, the device comprises:

[0092] The image acquisition module 210 is configured to acquire original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged.

[0093] The image reconstruction module 220 is configured to perform isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image.

[0094] The image segmentation module 230 is configured to segment the target MRI image to obtain a target segmentation image labeled with carpal bone image regions and triangular fibrocartilage complex (TFCC) image regions.

[0095] The stop point analysis module 240 is configured to perform edge extraction on the target segmentation image to obtain the stop point range of the TFCC on the carpal bone, and obtain the target quantitative index corresponding to the stop point range.

[0096] The device provided by the embodiment of the present application obtains original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged, performs isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image, performs segmentation on the target MRI image to obtain a target segmentation image labeled with a carpal bone image region and a triangular fibrocartilage complex (TFCC) image region, performs edge extraction on the target segmentation image to obtain a TFCC enthesis range on the carpal bone, and obtains a target quantitative index corresponding to the enthesis range. The present application can generate a three-dimensional high-resolution target MRI image with coherent structure, real texture and high credibility by fusing the original MRI images of the real multi-views through isotropic super-resolution reconstruction, and can solve the problems of large MRI layer thickness of the existing clinical wrist joint, less TFCC layers, low resolution and difficult-to-identify ligament structure, and has important significance for observation and pathological diagnosis of the wrist joint and the TFCC and the like. In addition, the present application can accurately depict the complex and fine three-dimensional anatomical structure of the TFCC and realize automatic and accurate correspondence between the fine three-dimensional anatomy and the radiology of the TFCC by efficiently and accurately extracting the enthesis range and the corresponding target quantitative index through comprehensive use of image processing, edge extraction, plane fitting and the like, so as to guide the diagnosis and treatment of TFCC injury and improve the clinical curative effect.

[0097] The wrist joint ligament image processing and analysis device described in the embodiment can be correspondingly referred to the wrist joint ligament image processing and analysis method provided by the present application described above, and will not be described in detail here.

[0098] Figure 3 An example of a schematic diagram of a physical structure of an electronic device is shown as Figure 3 The electronic device can include a processor 310, a communications interface 320, a memory 330 and a communications bus 340, wherein the processor 310, the communications interface 320 and the memory 330 can complete mutual communication through the communications bus 340. The processor 310 can invoke a logical instruction in the memory 330 to execute a wrist joint ligament image processing and analysis method, which includes:

[0099] Obtaining original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged;

[0100] Performing isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image;

[0101] Performing segmentation on the target MRI image to obtain a target segmentation image labeled with a carpal bone image region and a triangular fibrocartilage complex (TFCC) image region;

[0102] Edge extraction is performed on the target segmentation image to obtain a TFCC insertion point range on the carpal bone, and a target quantitative index corresponding to the insertion point range is obtained.

[0103] In addition, the logical instructions in the memory 330 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0104] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the wrist ligament image processing and analysis method provided by the above-mentioned methods, the method comprising:

[0105] Obtaining original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged;

[0106] Performing isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain target MRI images;

[0107] Segmenting the target MRI images to obtain a target segmentation image labeled with carpal bone image regions and triangular fibrocartilage complex (TFCC) image regions;

[0108] Performing edge extraction on the target segmentation image to obtain a TFCC insertion point range on the carpal bone, and obtaining a target quantitative index corresponding to the insertion point range.

[0109] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the wrist ligament image processing and analysis method provided by the above-mentioned methods, the method comprising:

[0110] Obtaining original magnetic resonance imaging (MRI) images of at least two different views of a wrist joint to be imaged;

[0111] perform isotropic super-resolution reconstruction on the original MRI images of the at least two different views to obtain a target MRI image;

[0112] perform segmentation on the target MRI image to obtain a target segmentation image labeled with a carpal bone image region and a triangular fibrocartilage complex, TFCC, image region;

[0113] perform edge extraction on the target segmentation image to obtain a TFCC enthesis range on the carpal bone, and obtain a target quantitative index corresponding to the enthesis range.

[0114] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0115] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0116] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A wrist joint ligament image processing and analysis method, characterized in that: The method comprises: Acquire original magnetic resonance imaging (MRI) images of at least two different views of the wrist joint to be imaged; perform isotropic super-resolution reconstruction on the original MRI images of at least two different views to obtain a target MRI image; Segmenting the target MRI image to obtain a target segmented image with a carpal bone image region and a triangular fibrocartilage complex (TFCC) image region marked; Performing edge extraction on the target segmentation image to obtain the end point range of the TFCC on the carpal bone, and performing plane fitting on the end point range based on the RANSAC algorithm; performing data statistics on the end point range after plane fitting to obtain the target quantitative index corresponding to the end point range; The performing edge extraction on the target segmented image to obtain the stop range of the TFCC on the carpal bone includes: Segmenting the carpal bone image region and the TFCC image region from the target segmented image; Processing the carpal bone image region based on a Sobel operator edge detection algorithm and a morphological dilation and erosion algorithm to obtain a first edge contour of the carpal bone image region; and Processing the TFCC image region based on a Sobel operator edge detection algorithm and a morphological dilation and erosion algorithm to obtain a second edge contour of the TFCC image region; Based on the first edge contour and the second edge contour, the intersection area of ​​the carpal bone image area and the TFCC image area is determined to obtain the end point range of the TFCC on the carpal bone.

2. The wrist joint ligament image processing and analysis method according to claim 1, characterized in that: The target MRI image is segmented to obtain a target segmented image with a carpal bone image region and a triangular fibrocartilage complex (TFCC) image region marked, including: The target MRI image is imported into 3D Slicer software, and the universal segmentation model SAM is called in the 3D Slicer software to interactively segment the target MRI image to obtain a target segmentation image with the carpal bone image area and the triangular fibrocartilage complex TFCC image area marked.

3. The wrist joint ligament image processing and analysis method according to claim 1, characterized in that: The wrist joint to be imaged includes a plurality of target segmented images in the Z-axis direction. Before edge extraction is performed on the target segmented images, the method further includes: Interpolation is performed on the plurality of target segmented images at alternate frames to supplement spatial information of the plurality of target segmented images in the Z-axis direction.

4. The wrist joint ligament image processing and analysis method according to claim 3, characterized in that: Also includes: Based on a joint smoothing algorithm, each target segmented image after interpolation of alternate frames is smoothed.

5. A wrist joint ligament image processing and analysis device, characterized in that: The device comprises: An image acquisition module, configured to acquire original magnetic resonance imaging (MRI) images of at least two different views of the wrist joint to be imaged; An image reconstruction module is used to perform isotropic super-resolution reconstruction on the original MRI images of at least two different views to obtain a target MRI image; an image segmentation module, configured to segment the target MRI image to obtain a target segmented image with a carpal bone image region and a triangular fibrocartilage complex (TFCC) image region marked thereon; An end-point analysis module is used to extract edges from the target segmented image to obtain the end-point range of the TFCC on the carpal bone, and perform plane fitting on the end-point range based on the RANSAC algorithm; perform data statistics on the end-point range after plane fitting to obtain the target quantitative index corresponding to the end-point range; The stop point analysis module is further used for: Segmenting the carpal bone image region and the TFCC image region from the target segmented image; Processing the carpal bone image region based on a Sobel operator edge detection algorithm and a morphological dilation and erosion algorithm to obtain a first edge contour of the carpal bone image region; and Processing the TFCC image region based on a Sobel operator edge detection algorithm and a morphological dilation and erosion algorithm to obtain a second edge contour of the TFCC image region; Based on the first edge contour and the second edge contour, the intersection area of ​​the carpal bone image area and the TFCC image area is determined to obtain the end point range of the TFCC on the carpal bone.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the wrist joint ligament image processing and analysis method according to any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the wrist joint ligament image processing and analysis method according to any one of claims 1 to 4 is implemented.

Images