Method and device for determining sagittal view angle of postoperative prosthesis based on knee joint lateral position sheet

Abstract

The invention provides a method and a device for determining a postoperative prosthesis sagittal view angle based on a knee joint lateral position film. The method comprises the following steps: acquiring a knee joint lateral position X-ray full-length film; key anatomical mark points of the femur, the tibia and the prosthesis are recognized based on the deep convolution model; fitting a central axis and an anatomical axis based on the key anatomical mark points; and calculating the sagittal view angle of the postoperative prosthesis through the central axis and the anatomical axis. According to the method, based on the knee joint lateral X-ray full-length film, the anatomical axes of the femur and the tibia are accurately fitted through an algorithm, and overall evaluation of the sagittal view angle of the prosthesis is achieved.

Description

Technical Field

[0001] This application relates to the technical field of medical image processing. Specifically, it relates to a method and device for determining the sagittal angle of a postoperative prosthesis based on a lateral knee radiograph. Background Art

[0002] Currently, the angle evaluation of the postoperative prosthesis implantation in the knee joint is an important parameter for judging the prosthesis stability, functional recovery, and revision surgery indication. The existing technology mainly relies on intraoperative mechanical guidance systems and local imaging examinations to determine the sagittal angle of the postoperative prosthesis.

[0003] However, the former operation is limited by the intraoperative environment and cannot reflect the long-term postoperative state; the latter has a limited imaging range, resulting in large measurement errors. Summary of the Invention

[0004] The problem solved by this application is that the existing methods for determining the sagittal angle of the postoperative prosthesis have large errors.

[0005] To solve the above problems, the first aspect of this application provides a method for determining the sagittal angle of a postoperative prosthesis based on a lateral knee radiograph, which includes:

[0006] Obtain a full-length lateral X-ray of the knee joint;

[0007] Based on a deep convolutional model, identify the key anatomical landmark points of the femur, tibia, and prosthesis;

[0008] Based on the key anatomical landmark points, fit the central axis and the anatomical axis;

[0009] Calculate the sagittal angle of the postoperative prosthesis through the central axis and the anatomical axis.

[0010] The second aspect of this application provides a manufacturing system for a method of determining the sagittal angle of a postoperative prosthesis based on a lateral knee radiograph, which includes:

[0011] An image acquisition module, which is used to obtain a full-length lateral X-ray of the knee joint;

[0012] A landmark point recognition module, which is used to identify the key anatomical landmark points of the femur, tibia, and prosthesis based on a deep convolutional model;

[0013] An axis fitting module, which is used to fit the central axis and the anatomical axis based on the key anatomical landmark points;

[0014] An angle calculation module, which is used to calculate the sagittal angle of the postoperative prosthesis through the central axis and the anatomical axis.

[0015] The third aspect of this application provides an electronic device, including: a memory and a processor; the memory can be configured to store a program, and the processor is coupled to the memory and is used to execute the program in the memory for:

[0016] Obtain a full-length lateral X-ray of the knee joint;

[0017] Identify the key anatomical landmark points of the femur, tibia, and prosthesis based on a deep convolutional model;

[0018] Based on the key anatomical landmark points, fit the central axis and the anatomical axis;

[0019] Calculate the sagittal angle of the postoperative prosthesis through the central axis and the anatomical axis.

[0020] The fourth aspect of the present application provides a computer-readable storage medium, on which a computer program is stored, and the program is executed by a processor to implement the method for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph described above.

[0021] In the present application, based on the full-length lateral X-ray of the knee joint, the anatomical axes of the femur and tibia are accurately fitted through an algorithm to achieve an overall evaluation of the sagittal angle of the prosthesis. Description of the Drawings

[0022] Figure 1 It is a flowchart of the method for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph according to an embodiment of the present application;

[0023] Figure 2 It is an architecture diagram of the point recognition model for the method for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph according to an embodiment of the present application;

[0024] Figure 3 It is an architecture diagram of the device for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph according to an embodiment of the present application;

[0025] Figure 4 It is an architecture diagram of the electronic device according to an embodiment of the present application. Detailed Description of the Embodiments

[0026] To make the above objects, features, and advantages of the present application more obvious and understandable, the following detailed description of the specific embodiments of the present application is provided in conjunction with the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0027] It should be noted that unless otherwise specified, the technical terms or scientific terms used in the present application should have the ordinary meaning understood by those skilled in the art to which the present application belongs.

[0028] Currently, the angular evaluation of the prosthesis implantation after knee joint surgery is an important parameter for judging the prosthesis stability, functional recovery and the indication of revision surgery. The existing technologies mainly rely on the following methods:

[0029] Intraoperative mechanical guidance system: Monitor the installation angle of the prosthesis through an intraoperative locator.

[0030] Disadvantages: The operation is limited by the intraoperative environment, and the accuracy of the data may be affected by the soft tissue tension state or anatomical abnormalities, and it cannot reflect the long-term postoperative state.

[0031] Imaging examination:

[0032] Local lateral X-ray film: Evaluate the installation angle of the prosthesis through the local image of the knee joint.

[0033] Disadvantages: The local image range is limited, and it is difficult to obtain the precise relationship between the prosthesis and the overall force lines of the femur and tibia, resulting in large measurement errors.

[0034] The embodiment of the present application provides the method for determining the sagittal angle of the postoperative prosthesis based on the lateral knee film as described above. The specific scheme of this method is Figures 1 - 4 shown. This method can be executed by a device for determining the sagittal angle of the postoperative prosthesis based on the lateral knee film. The device for determining the sagittal angle of the postoperative prosthesis based on the lateral knee film can be integrated in electronic devices such as computers, servers, computers, server clusters, data centers, etc. Combining Figure 1 shown, wherein, the method for determining the sagittal angle of the postoperative prosthesis based on the lateral knee film includes:

[0035] S101, Obtain the full-length lateral X-ray film of the knee joint;

[0036] In the present application, the full-length lateral X-ray film of the knee joint is an image taken from the side of the knee joint by X-rays, aiming to obtain clear images of the femur, tibia, patella and knee joint prosthesis (such as artificial knee joint). Such images can show the structural relationship of the knee joint, the alignment of the bones, the width of the joint space, as well as the position and angle of the prosthesis.

[0037] S102, Identify the key anatomical landmark points of the femur, tibia and prosthesis based on the deep convolutional model;

[0038] S103, Fit the central axis and the anatomical axis based on the key anatomical landmark points;

[0039] S104, Calculate the sagittal angle of the postoperative prosthesis through the central axis and the anatomical axis.

[0040] In the present application, based on the full-length lateral X-ray film of the knee joint, the anatomical axes of the femur and tibia are accurately fitted through algorithms to achieve the overall evaluation of the sagittal angle of the prosthesis.

[0041] In this application, an AI algorithm is introduced to automatically identify the prosthesis and bony landmark points, significantly improving the efficiency and accuracy of measurement.

[0042] When measuring the flexion and extension angles of the knee joint in the prior art, the following main deficiencies exist:

[0043] Inability to evaluate the overall sagittal alignment: The local X-ray film cannot cover the entire lengths of the femur and tibia simultaneously, making it difficult to reflect the position and angle of the prosthesis in the overall sagittal alignment.

[0044] High cost and complexity: Although CT technology has high precision, it has a large radiation dose and high cost, making it difficult to be applied in routine follow-up or large-scale screening.

[0045] Reliance on subjective experience: Most of the existing X-ray evaluation methods rely on manual measurement of the angular relationship between the prosthesis and the bone structure, which is affected by the experience of radiologists and results in low consistency of the results.

[0046] Compared with the prior art, the present invention has the following significant differences:

[0047] Full-length image coverage: It provides the overall relationship of the femur, tibia and prosthesis, overcoming the range limitation of local images.

[0048] Automated analysis: Through AI technology, anatomical landmark point recognition and angle calculation are realized, reducing manual errors and improving consistency.

[0049] Low-cost applicability: Compared with CT technology, the present invention only requires conventional X-ray imaging equipment, has a significant cost advantage, and is suitable for postoperative follow-up and large-scale applications.

[0050] In one embodiment, after calculating the sagittal angle of the postoperative prosthesis, a graphical report including the anatomical axis, the central axis of the prosthesis and the angle calculation result is generated through a computer display module, and a diagnostic report in PDF format is output for clinical reference.

[0051] The specific processing process includes: loading and displaying the lateral X-ray image of the patient's knee joint as the background, superimposing and displaying the anatomical axes of the femur and tibia on the image, marking their directions and positions in a straight line form, displaying the central axis of the prosthesis, and distinguishing the central axis of the prosthesis and the anatomical axis with different colors; clearly marking the sagittal angle of the femoral prosthesis, the sagittal angle of the tibial prosthesis and the sagittal angle of the postoperative prosthesis on the image, and representing their geometric relationship with intuitive arrows or arcs.

[0052] Among them, in the specific processing process, if there is an error range or a preset reference value range, it can be shown in the report by numerical or color differentiation (for example, green represents the normal range and red represents the abnormal angle); below or on the right side of the graphical report, key data and conclusions related to the diagnosis are displayed, including: the specific numerical value of the angle calculation result, whether it meets the ideal value range set clinically, and suggestions or warnings (such as the need for further examination or intervention).

[0053] In one embodiment, the full-length lateral X-ray of the knee joint covers the entire lengths of the femur and tibia and the complete form of the prosthesis.

[0054] In this application, the starting part of the femur is the femoral head, and the starting part of the tibia is the condyle of the tibia. The full-length film can show the complete structure from the upper end of the femur to the lower end of the tibia, ensuring a comprehensive evaluation of the knee joint.

[0055] In this application, the complete form of the prosthesis includes the size, shape, installation position of the prosthesis, and the connection state with the bones (femur and tibia). Its complete form can be obtained by taking an X-ray that includes the entire prosthesis, that is, the X-ray needs to include the entire prosthesis.

[0056] In one embodiment, the key anatomical landmark points include: the starting point of the central axis of the femoral prosthesis, the connection point of the distal femoral prosthesis, the starting point of the tibial prosthesis, and the distal prosthesis connection point.

[0057] In this application, the starting point of the central axis of the femoral prosthesis refers to the access position of the femoral prosthesis at the femoral end during knee joint replacement surgery.

[0058] In this application, the connection point of the distal femoral prosthesis refers to the connection point of the femoral prosthesis at the distal end of the knee joint (i.e., the end close to the patella) with the femur. It is usually the place where the femoral prosthesis and the femur are joined, marking the end of the connection between the prosthesis and the femur.

[0059] In this application, the starting point of the tibial prosthesis refers to the access position of the prosthesis at the tibial part during knee joint replacement surgery.

[0060] In this application, the distal tibial prosthesis connection point refers to the connection point of the prosthesis at the distal end of the tibia (i.e., the end close to the ankle) with the tibia. It marks the end of the access of the prosthesis in the tibia and the contact area between the prosthesis and the tibia.

[0061] In one embodiment, the central axis is the central axis of the prosthesis; the anatomical axes include the femoral anatomical axis and the tibial anatomical axis.

[0062] In this application, the central axis of the prosthesis refers to the assumed center line of the prosthesis (such as a knee joint replacement prosthesis) along its longitudinal axis. The central axis of the prosthesis is usually the line connecting the femoral part to the tibial part of the prosthesis, which is the "geometric center" axis of the prosthesis.

[0063] In this application, the femoral anatomical axis refers to the central line from the femoral head to the distal femoral condyles, representing the natural longitudinal axis of the femur. The femoral anatomical axis generally reflects the physiological curvature and anatomical structure characteristics of the femur.

[0064] In this application, the tibial anatomical axis refers to the central line from the tibial plateau to the distal tibial condyles, representing the natural longitudinal axis of the tibia, and reflecting the physiological structure and symmetry of the tibia.

[0065] In this application, the starting point of the central axis of the femoral prosthesis fits with the connecting point of the distal femoral prosthesis to obtain the central axis of the femoral prosthesis.

[0066] In this application, the starting point of the tibial prosthesis and the connecting point of the distal prosthesis are fitted to obtain the central axis of the tibial prosthesis.

[0067] In this application, the femoral anatomical axis is the central axis of the natural anatomical structure of the femur. It is the long axis of the femur and is determined by the anatomical landmark points of the femur. Common landmark points include the upper end of the femur (the position of the femoral head) and the lower end of the femur (the femoral condyles).

[0068] In this application, the tibial anatomical axis is similar to the femoral anatomical axis. It represents the natural long axis of the tibia and is determined by the anatomical landmark points of the tibia. The landmark points are usually the condyles at the upper end of the tibia and the lower condyles or intercondylar area at the lower end of the tibia.

[0069] In one embodiment, calculating the postoperative prosthesis sagittal angle through the central axis and the anatomical axis includes:

[0070] Calculating the femoral prosthesis sagittal angle based on the femoral anatomical axis and the central axis of the prosthesis;

[0071] Calculating the tibial prosthesis sagittal angle based on the tibial anatomical axis and the central axis of the prosthesis;

[0072] Calculating the postoperative prosthesis sagittal angle based on the femoral prosthesis sagittal angle and the tibial prosthesis sagittal angle.

[0073] In this application, by calculating the included angle between the femoral anatomical axis and the central axis of the prosthesis, the femoral prosthesis sagittal angle is obtained.

[0074] In this application, the included angle between the tibial anatomical axis and the central axis of the prosthesis is calculated to obtain the tibial prosthesis sagittal angle.

[0075] In this application, the postoperative prosthesis sagittal angle is generally the weighted average of the femoral prosthesis sagittal angle and the tibial prosthesis sagittal angle. It can also be determined by other comprehensive methods. The specific determination method is not elaborated in this application.

[0076] In this application, a comprehensive sagittal angle is calculated through weighted averaging or other mathematical methods from these two angles to determine whether the overall positioning of the prosthesis meets the postoperative expected standard.

[0077] In one implementation, in combination with Figure 2 As shown, the key anatomical landmark points of the femur, tibia, and prosthesis identified based on the deep convolutional model include:

[0078] Input the full-length lateral X-ray film of the knee joint into the feature extraction structure to obtain a feature extraction map;

[0079] Perform splicing residual processing on the feature extraction map to obtain a residual feature map;

[0080] Perform convolutional processing on the feature extraction map, splice it with the residual feature map, and then perform point recognition processing to obtain the key anatomical landmark points.

[0081] In this application, in combination with Figure 2 As shown, the full-length lateral X-ray film of the knee joint is used as the input image and processed through the feature extraction structure (CNN). The feature extraction structure extracts low-level and high-level features (such as edges, textures, shapes, etc.) from the image through learned convolutional kernels and generates a feature extraction map.

[0082] In this application, performing splicing residual processing on the feature extraction map is to enhance the expression ability of the model and use different convolutional branches to capture different scales and features. The generation process of the residual feature map can effectively improve the training efficiency of the network and alleviate the problem of gradient disappearance.

[0083] In this application, the feature extraction map and the residual feature map are further processed through convolutional operations. After combining the two, point recognition processing is performed. This step can locate the key anatomical landmark points of the femur, tibia, and prosthesis through a specific point recognition module (such as a fully connected layer, regression network, etc.).

[0084] In one implementation, in combination with Figure 2 As shown, the process of performing splicing residual processing on the feature extraction map to obtain a residual feature map includes:

[0085] Perform first-branch convolutional processing on the feature extraction map to obtain a first-branch map;

[0086] Perform second-branch convolutional processing on the feature extraction map to obtain a second-branch map;

[0087] Perform third-branch convolutional processing on the feature extraction map to obtain a third-branch map;

[0088] Splice the first-branch map, the second-branch map, and the third-branch map and then perform convolutional processing to obtain a spliced map;

[0089] After adding the stitched image and the feature extraction image and activating through Relu, a residual feature map is obtained.

[0090] In this application, in combination with Figure 2 As shown, the feature extraction image is processed through a convolution operation to generate a first branch image. The first branch convolution can capture local information in the feature extraction image, such as low-level texture or edge features.

[0091] Similarly, the feature extraction image can also be further processed through a second branch convolution to obtain a second branch image. The second branch convolution can capture different features or scales.

[0092] The third branch convolution processing can further extract different-level or higher-level features, thereby generating a third branch image.

[0093] The output images of the three branches are stitched together to synthesize a multi-dimensional stitched image, containing multi-level features from different branches. Then, the information in the stitched image is further fused through convolution processing to obtain more advanced features.

[0094] Finally, the stitched image is added to the original feature extraction image for residual connection (skip connection). This connection method helps to accelerate the training of the model and alleviate the problem of gradient disappearance. After the addition operation, through the ReLU activation function, the final residual feature map is obtained.

[0095] In this application, the first branch is a 1×1 convolution (dilated convolution) with a dilation rate of 1, the second branch is a 3×3 convolution (dilated convolution) with a dilation rate of 3, and the third branch is a 5×5 convolution (dilated convolution) with a dilation rate of 5; among them, the 5×5 convolution kernel can capture a larger receptive field range, and is more suitable for processing more complex local features compared to the 3×3 convolution kernel; the dilation rate is 5, that is, there will be an interval of 4 pixels between every two adjacent elements of the convolution kernel; the receptive field of the dilated convolution kernel will increase, but the actual number of parameters remains unchanged.

[0096] In this application, dilated convolutions with corresponding dilation rates are set, so that the range (receptive field) covered by the convolution kernel becomes larger, but the same weight parameters are still used. In the case of a deeper number of feature layers, the receptive field is expanded without increasing the convolution kernel parameters.

[0097] In this application, different branches are set corresponding to different receptive fields, so that while generating feature maps with different receptive fields, the number of weight parameters of the feature maps is also maintained the same, so that they can be directly stitched together to improve the processing efficiency of the entire model.

[0098] In this application, the measurement accuracy of this solution was verified in the postoperative images of real patients. The results showed that:

[0099] Compared with the sagittal angle of the prosthesis measured by CT, the error was less than ±2°, meeting the clinical requirements.

[0100] In the repeatability test, the consistency (ICC value) of the measurement results for different patients was greater than 0.95, showing good stability and reliability.

[0101] The embodiment of this application provides a device for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph, which is used to execute the method for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph described above. The following is a detailed description of the device for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph.

[0102] As Figure 3 shown, the device for determining the sagittal angle of the postoperative prosthesis based on the lateral knee radiograph includes:

[0103] An image acquisition module 101, which is used to acquire a full-length lateral X-ray film of the knee joint;

[0104] A landmark recognition module 102, which is used to recognize the key anatomical landmarks of the femur, tibia, and prosthesis based on a deep convolutional model;

[0105] An axis fitting module 103, which is used to fit the central axis and the anatomical axis based on the key anatomical landmarks;

[0106] An angle calculation module 104, which is used to calculate the sagittal angle of the postoperative prosthesis through the central axis and the anatomical axis.

[0107] In one implementation, the full-length lateral X-ray film of the knee joint covers the entire lengths of the femur and tibia and the complete form of the prosthesis.

[0108] In one implementation, the key anatomical landmarks include: the starting point of the central axis of the femoral prosthesis, the connection point of the distal femoral prosthesis, the starting point of the tibial prosthesis, and the connection point of the distal prosthesis.

[0109] In one implementation, the central axis is the central axis of the prosthesis; the anatomical axis includes the femoral anatomical axis and the tibial anatomical axis.

[0110] In one implementation, the angle calculation module 104 is further used for:

[0111] Calculating the sagittal angle of the femoral prosthesis based on the femoral anatomical axis and the central axis of the prosthesis; calculating the sagittal angle of the tibial prosthesis based on the tibial anatomical axis and the central axis of the prosthesis; calculating the sagittal angle of the postoperative prosthesis based on the sagittal angle of the femoral prosthesis and the sagittal angle of the tibial prosthesis.

[0112] In one embodiment, the landmark recognition module 102 is further configured to:

[0113] Input the full-length lateral X-ray film of the knee joint into the feature extraction structure to obtain a feature extraction map; perform stitching residual processing on the feature extraction map to obtain a residual feature map; perform convolution processing on the feature extraction map, and after stitching with the residual feature map, perform point recognition processing to obtain key anatomical landmarks.

[0114] In one embodiment, the landmark recognition module 102 is further configured to:

[0115] Perform first-branch convolution processing on the feature extraction map to obtain a first-branch map; perform second-branch convolution processing on the feature extraction map to obtain a second-branch map; perform third-branch convolution processing on the feature extraction map to obtain a third-branch map; perform convolution processing after stitching the first-branch map, the second-branch map, and the third-branch map to obtain a stitched map; after adding the stitched map and the feature extraction map, activate through Relu to obtain a residual feature map.

[0116] The device for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film provided in the above embodiments of the present application has a corresponding relationship with the method for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film provided in the embodiments of the present application. Therefore, the specific content in this system has a corresponding relationship with the method for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film. The specific content can refer to the records in the method for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film, and will not be elaborated herein.

[0117] The device for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film provided in the above embodiments of the present application and the method for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application programs stored therein.

[0118] The internal functions and structures of the device for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film are described above. As Figure 4 shown, in practice, the device for determining the postoperative prosthesis sagittal angle based on the lateral knee joint film can be implemented as an electronic device, including: a memory 301 and a processor 303.

[0119] The memory 301 can be configured to store programs.

[0120] In addition, the memory 301 can also be configured to store various other data to support operations on the electronic device. Examples of these data include instructions for any application program or method for operating on the electronic device, contact data, phone book data, messages, pictures, videos, etc.

[0121] The memory 301 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc. The processor 303, coupled to the memory 301, is configured to execute a program in the memory 301 for:

[0122] Obtain a full-length lateral X-ray film of the knee joint;

[0123] Identify key anatomical landmark points of the femur, tibia, and prosthesis based on a deep convolutional model;

[0124] Fit the central axis and the anatomical axis based on the key anatomical landmark points;

[0125] Calculate the sagittal angle of the postoperative prosthesis through the central axis and the anatomical axis.

[0126] In one embodiment, the full-length lateral X-ray film of the knee joint covers the entire lengths of the femur and tibia and the complete form of the prosthesis.

[0127] In one embodiment, the key anatomical landmark points include: the starting point of the central axis of the femoral prosthesis, the connecting point of the distal femoral prosthesis, the starting point of the tibial prosthesis, and the connecting point of the distal prosthesis.

[0128] In one embodiment, the central axis is the central axis of the prosthesis; the anatomical axis includes the femoral anatomical axis and the tibial anatomical axis.

[0129] In one embodiment, the processor 303 is further configured to:

[0130] Calculate the sagittal angle of the femoral prosthesis based on the femoral anatomical axis and the central axis of the prosthesis; calculate the sagittal angle of the tibial prosthesis based on the tibial anatomical axis and the central axis of the prosthesis; calculate the sagittal angle of the postoperative prosthesis based on the sagittal angle of the femoral prosthesis and the sagittal angle of the tibial prosthesis.

[0131] In one embodiment, the processor 303 is further configured to:

[0132] Input the full-length lateral X-ray film of the knee joint into a feature extraction structure to obtain a feature extraction map; perform splicing and residual processing on the feature extraction map to obtain a residual feature map; perform convolutional processing on the feature extraction map, splice it with the residual feature map, and then perform point recognition processing to obtain the key anatomical landmark points.

[0133] In one embodiment, the processor 303 is further configured to:

[0134] Perform first-branch convolution processing on the feature extraction graph to obtain the first-branch graph; perform second-branch convolution processing on the feature extraction graph to obtain the second-branch graph; perform third-branch convolution processing on the feature extraction graph to obtain the third-branch graph; splice the first-branch graph, the second-branch graph, and the third-branch graph and then perform convolution processing to obtain the spliced graph; add the spliced graph and the feature extraction graph and then activate through Relu to obtain the residual feature graph.

[0135] In this application, Figure 4 only some components are schematically shown, and it does not mean that the electronic device only includes Figure 4 the components shown.

[0136] The electronic device provided in this embodiment and the method for determining the postoperative prosthesis sagittal angle based on the knee lateral radiograph provided in the embodiment of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application programs stored therein.

[0137] Those skilled in the art should understand that the embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0138] This application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, causing a series of operational steps to be performed on the computer or other programmable apparatus to generate a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process Figure 1 one process or a plurality of processes and / or blocks Figure 1 steps for implementing the functions specified in one block or a plurality of blocks.

[0140] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory. The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (Flash RAM). Memory is an example of computer-readable media.

[0141] This application also provides a computer-readable storage medium corresponding to the method for determining the sagittal angle of a postoperative prosthesis based on a lateral knee radiograph provided in the foregoing embodiment. A computer program (i.e., a program product) is stored thereon. When the computer program is run by a processor, it will execute the method for interactive image analysis assistance for 3D aerial imaging provided in any of the foregoing embodiments.

[0142] Computer-readable media includes both permanent and non-permanent, removable and non-removable media and can be implemented by any method or technology for storing information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transitory media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0143] The computer-readable storage medium provided by the above embodiments of this application and the method for interactive image analysis assistance for 3D aerial imaging provided by the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application programs stored thereon.

[0144] Note that in the specification provided herein, a large number of specific details are described. However, it can be understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this specification.

[0145] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements not only includes those elements but also other elements not expressly listed, or elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising said element.

[0146] The above are only the embodiments of the present application and are not used to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for determining the sagittal angle of a postoperative prosthesis based on a lateral knee radiograph, characterized in that: include: Obtain a full-length lateral knee X-ray; Identify key anatomical landmarks of the femur, tibia, and prosthesis based on a deep convolutional model; Based on the key anatomical landmarks, fitting the central axis and the anatomical axis; The postoperative sagittal angle of the prosthesis is calculated through the central axis and the anatomical axis.

2. The method for determining the sagittal angle of a postoperative prosthesis based on a lateral view of the knee joint according to claim 1, characterized in that: The full-length lateral X-ray film of the knee joint covers the full length of the femur and tibia and the complete shape of the prosthesis.

3. The method for determining the sagittal angle of a postoperative prosthesis based on a lateral view of the knee joint according to claim 1, characterized in that: The key anatomical landmarks include: the starting point of the femoral prosthesis mid-axis, the femoral distal prosthesis connection point, the tibial prosthesis starting point and the distal prosthesis connection point.

4. The method for determining the sagittal angle of a postoperative prosthesis based on a lateral knee radiograph according to any one of claims 1 to 3, characterized in that: The central axis is the central axis of the prosthesis; the anatomical axis includes the femoral anatomical axis and the tibia anatomical axis.

5. The method for determining the sagittal angle of a postoperative prosthesis based on a lateral view of the knee joint according to claim 4, characterized in that: The calculating of the postoperative sagittal angle of the prosthesis through the central axis and the anatomical axis includes: The sagittal angle of the femoral prosthesis was calculated based on the femoral anatomical axis and the prosthesis midline; The sagittal angle of the tibial prosthesis was calculated based on the anatomical axis of the tibia and the midline axis of the prosthesis; The postoperative sagittal angle of the prosthesis was calculated based on the sagittal angle of the femoral component and the sagittal angle of the tibial component.

6. The method for determining the sagittal angle of a postoperative prosthesis based on a lateral knee radiograph according to any one of claims 1 to 3, characterized in that: The deep convolutional model is used to identify key anatomical landmarks of the femur, tibia, and prosthesis, including: Input the full-length lateral X-ray film of the knee joint into the feature extraction structure to obtain a feature extraction graph; Perform splicing residual processing on the feature extraction map to obtain a residual feature map; The feature extraction map is convolved and concatenated with the residual feature map, and then point recognition is performed to obtain key anatomical landmarks.

7. The method for determining the sagittal angle of a postoperative prosthesis based on a lateral view of the knee joint according to claim 6, characterized in that: The step of performing splicing residual processing on the feature extraction graph to obtain a residual feature graph includes: Performing a first branch convolution process on the feature extraction graph to obtain a first branch graph; Performing a second branch convolution process on the feature extraction graph to obtain a second branch graph; Performing a third branch convolution process on the feature extraction graph to obtain a third branch graph; The first branch graph, the second branch graph, and the third branch graph are concatenated and then convolved to obtain a concatenated graph; After adding the concatenated image and the feature extraction image, the residual feature map is obtained through Relu activation.

8. A device for determining the sagittal angle of a prosthesis after surgery based on a lateral view of a knee joint, characterized in that: include: An image acquisition module, which is used to obtain a full-length lateral X-ray film of the knee joint; A landmark recognition module, which is used to identify key anatomical landmarks of the femur, tibia, and prosthesis based on a deep convolutional model; An axis fitting module, which is used to fit the central axis and the anatomical axis based on the key anatomical landmarks; The angle calculation module is used to calculate the sagittal angle of the postoperative prosthesis through the central axis and the anatomical axis.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store programs; The processor, coupled to the memory, is configured to execute the program to: Obtain a full-length lateral knee X-ray; Identify key anatomical landmarks of the femur, tibia, and prosthesis based on a deep convolutional model; Based on the key anatomical landmarks, fitting the central axis and the anatomical axis; The postoperative sagittal angle of the prosthesis is calculated through the central axis and the anatomical axis.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by the processor to implement the method for determining the sagittal angle of a postoperative prosthesis based on a lateral knee joint radiograph as described in any one of claims 1-6.

Images