WO2024060544A1

WO2024060544A1 - Method and system for intelligent design of positioning apparatus for knee joint with complex synostosis

Info

Publication number: WO2024060544A1
Application number: PCT/CN2023/082727
Authority: WO
Inventors: 张逸凌; 刘星宇
Original assignee: 北京长木谷医疗科技有限公司; 张逸凌
Priority date: 2022-09-21
Filing date: 2023-03-21
Publication date: 2024-03-28
Also published as: CN115381553A; CN115381553B

Abstract

Provided is a method for intelligent design of a positioning apparatus for a knee joint with complex synostosis, comprising: performing recognition processing on a medical image of a knee joint with synostosis by means of a key point recognition model to acquire a plurality of key points of the knee joint with synostosis (S120); performing recognition processing on the medical image of the knee joint with synostosis by means of a fitting area recognition model to acquire two fitting parts of the knee joint with synostosis (S130); and generating, on the basis of the plurality of key points of the knee joint with synostosis and the two fitting parts of the knee joint with synostosis, a positioning apparatus for the knee joint with synostosis (S140). In this way, the design of the positioning apparatus does not need to manually or empirically mark parameters, thereby improving the efficiency of manufacturing the positioning apparatus and the fit extent of the positioning apparatus and the knee joint with synostosis, and thus improving the accuracy when the positioning apparatus is used for osteotomy.

Description

Intelligent positioning device design method and system for complex bony fusion knee joint

Cross-references to related applications

This application claims priority to the Chinese patent application with application number 202211148280.1 submitted on September 21, 2022, titled "Design Method and System for Intelligent Positioning Devices for Complex Bone Fusion Knee Joints", which is fully incorporated by reference. Enter this article.

Technical field

The present application relates to the field of computer technology, specifically, to a design method and system for an intelligent positioning device for complex bony fusion knee joints.

Background technique

Knee joint bony fusion is a type of orthopedic disease. The main causes of knee joint bony fusion include ankylosing spondylitis, rheumatoid arthritis, bone tuberculosis, synovectomy, etc. The current implementation of surgery relies on the experience of the surgeon and the assistance of an osteotomy guide plate to complete the osteotomy. The existing osteotomy guide design method relies on manual experience to manually mark the three-dimensional knee joint model to mark some parameters for designing the osteotomy guide, and manufacture the osteotomy guide based on the design parameters.

However, when realizing the inventive concept of the present application, the inventor found that there are at least the following technical problems in the related art: the efficiency of making the osteotomy guide plate using the above method is low, and the matching degree between the osteotomy guide plate made by this method and the bony fusion knee joint Not high, which reduces the accuracy of osteotomy using this osteotomy guide plate.

Contents of the invention

The main purpose of this application is to provide a design method and system for an intelligent positioning device for complex bony fusion knee joints that can improve the efficiency of guide plate design.

In order to achieve the above purpose, according to one aspect of the present application, a design method of an intelligent positioning device for a complex bony fusion knee joint is provided, which uses a three-dimensional reconstruction model to analyze the bony fusion knee joint. The medical image is segmented, and a three-dimensional image of the bony fusion knee joint is generated based on the segmentation results; the medical image of the bony fusion knee joint is recognized and processed through a key point recognition model, and a multi-dimensional image of the bony fusion knee joint is obtained. Key points; identify and process the medical image of the bony fusion knee joint by fitting a region recognition model to obtain two fitting parts of the bony fusion knee joint; based on the multiple bony fusion knee joints Key points and two fitting parts of the bony fusion knee joint are generated to generate a positioning device for the bony fusion knee joint; multiple features of the bony fusion knee joint are displayed on the three-dimensional image of the bony fusion knee joint. Key points, two fitting parts of the bony fusion knee joint, and a positioning device of the bony fusion knee joint.

Optionally, performing segmentation processing on the medical image of the bony fusion knee joint through the three-dimensional reconstruction model, and generating the three-dimensional image of the bony fusion knee joint based on the segmentation results includes: The medical image of the knee joint is input into the UNet model, and the medical image of the bony fusion knee joint is roughly segmented through the UNet model to obtain a coarse segmentation image; the coarse segmentation image is input into the PonitRend model, and the coarse segmentation image is obtained through the PonitRend model. The model performs pixel-level segmentation on the rough segmentation image, and generates a three-dimensional image of the bony fusion knee joint based on the pixel-level segmentation results.

Optionally, identifying and processing the medical image of the bony fusion knee joint through the key point recognition model to obtain multiple key points of the bony fusion knee joint includes: converting the bony fusion knee joint into The medical image of the joint is input into the Hourglass model, and the features in the medical image of the bony fused knee joint are detected through the Hourglass model to obtain the center point of the femoral head, the apex of the intercondylar fossa, and the ankle joint of the bony fused knee joint. center point, the lowest point of the medial and lateral tibial plateau, and the most distal point of the femoral condyle.

Optionally, the step of identifying and processing the medical image of the bony fusion knee joint by fitting a region recognition model to obtain the two fitting parts of the bony fusion knee joint includes: The medical image of the joint is input into the deeplabv3+ model, and the medical image of the bony fusion knee joint is identified and processed through the deeplabv3+ model to obtain the climbing area of the femoral anterior condyle and the tibial tubercle deviation of the bony fusion knee joint. Upper medial area.

Optionally, generating a positioning device for a bony fusion knee joint based on multiple key points of the bony fusion knee joint and two fitting sites of the bony fusion knee joint includes: according to the bony fusion knee joint Fusion of the center point of the femoral head, the apex of the intercondylar fossa, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the most distal point of the femoral condyle of the knee joint are used to determine the joint line dart groove, femoral dart groove, and tibial dart groove respectively. The position information of the positioning device; according to the femoral anterior condyle of the bony fusion knee joint The climbing area and the upper medial area of the tibial tubercle fit two fitting areas of the positioning device. The two fitting areas include the femoral fitting area and the tibial fitting area; based on the joint line dart The position information of the groove, the femoral dart groove, and the tibial dart groove in the positioning device and the femoral fitting area and tibial fitting area of the positioning device generate the positioning device of the bony fusion knee joint.

Optionally, the joint line dart groove is determined respectively based on the center point of the femoral head, the apex of the intercondylar fossa, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the most distal point of the femoral condyle of the bony fusion knee joint. The position information of the femoral dart groove and the tibial dart groove on the positioning device includes: determining the position of the joint line dart groove on the positioning device based on the lowest point of the internal and external tibial platform and the most distal point of the femoral condyle. Position information, the number of the joint line dart grooves is 1; according to the center point of the femoral head and the apex of the intercondylar fossa, the position information of the femoral dart grooves in the positioning device is determined, the femoral dart grooves The number of the tibial dart grooves is 2; the position information of the tibial dart grooves in the positioning device is determined based on the apex of the intercondylar fossa and the center point of the ankle joint, and the number of the tibial dart grooves is 2.

Optionally, the three-dimensional image of the bony fusion knee joint displays multiple key points of the bony fusion knee joint, two fitting parts of the bony fusion knee joint, and the bone The positioning device of the bony fusion knee joint includes: the femoral head in the three-dimensional image of the bony fusion knee joint displays the femoral head center point; the tibia in the three-dimensional image of the bony fusion knee joint displays the ankle joint. joint center point; showing the apex of the intercondylar fossa, the lowest point of the internal and external tibial plateau, and the most distal point of the femoral condyle at the junction of the femur and the tibia in the three-dimensional image of the bony fused knee joint; and at In the three-dimensional image of the bony fusion knee joint, the climbing area of the anterior femoral condyle and the upper and medial area of the tibial tubercle display the matching positioning device of the bony fusion knee joint.

According to another aspect of the present application, the present application also provides a system for designing an intelligent positioning device for a complex bony fusion knee joint, comprising: a three-dimensional reconstruction module, configured to segment and process a medical image of the bony fusion knee joint through a three-dimensional reconstruction model, and generate a three-dimensional image of the bony fusion knee joint based on the segmentation result; a key point recognition module, configured to recognize and process the medical image of the bony fusion knee joint through a key point recognition model to obtain multiple key points of the bony fusion knee joint; a fitting part recognition module, configured to recognize and process the medical image of the bony fusion knee joint through a fitting area recognition model to obtain two fitting parts of the bony fusion knee joint; a positioning device generation module, configured to recognize and process the medical image of the bony fusion knee joint based on the multiple key points and The two fitting parts of the bony fusion knee joint generate a positioning device for the bony fusion knee joint; a display module is configured to display multiple key points of the bony fusion knee joint, the two fitting parts of the bony fusion knee joint, and the positioning device for the bony fusion knee joint on a three-dimensional image of the bony fusion knee joint.

According to another aspect of the present application, the present application also provides a computer device and a computer-readable storage medium.

A computer device includes a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the computer program, the steps in each of the above method embodiments are implemented.

A computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by a processor, the steps of any one of the above methods are implemented.

This application provides an intelligent positioning device design method and system for complex bony fusion knee joints. This method can identify and process medical images of bony fusion knee joints through key point recognition models, and obtain multiple images of bony fusion knee joints. key points, and the medical image of the bony fusion knee joint is recognized and processed by fitting the area recognition model to obtain two fitting parts of the bony fusion knee joint, and based on multiple key points and bones of the bony fusion knee joint The two fitting parts of the knee joint are fused to generate a positioning device for the bony fusion knee joint. Designing the positioning device in this way does not require manual marking of parameters based on manual experience. This can improve the efficiency of making the positioning device and improve the positioning device. The matching degree with the bony fusion knee joint can thereby improve the accuracy of osteotomy using this positioning device.

Description of the drawings

The accompanying drawings, which constitute a part of this application, are included to provide a further understanding of the application so that other features, objects and advantages of the application will become apparent. The drawings and descriptions of the schematic embodiments of the present application are used to explain the present application and do not constitute an improper limitation of the present application. In the attached picture:

Figure 1 schematically shows a flow chart of a design method for an intelligent positioning device for complex bony fusion knee joints according to an embodiment of the present application;

Figure 2 schematically shows a schematic diagram of a medical image of a bony fused knee joint segmented by a three-dimensional reconstruction model according to an embodiment of the present application;

Figure 3 schematically shows a schematic diagram of a key point recognition model segmenting a medical image of a bony fusion knee joint according to an embodiment of the present application;

Figure 4 schematically shows a schematic diagram of a fitted region recognition model segmenting a medical image of a bony fusion knee joint according to an embodiment of the present application;

Figure 5 schematically shows the structural diagram of an intelligent positioning device for bony fusion knee joint according to an embodiment of the present application;

Figure 6 schematically shows the structural diagram of an intelligent positioning device for bony fusion knee joint according to another embodiment of the present application;

Figure 7 schematically shows a front view of an osteotomy performed by a positioning device in a three-dimensional image of a bony fused knee joint according to an embodiment of the present application;

Figure 8 is a side view schematically showing an osteotomy performed by a positioning device in a three-dimensional image of a bony fusion knee joint according to an embodiment of the present application;

FIG9 schematically shows a block diagram of a system for designing an intelligent positioning device for a complex bone fusion knee joint according to an embodiment of the present application;

Figure 10 schematically shows the internal structure diagram of a computer device according to an embodiment of the present application.

Detailed ways

In order to enable those in the technical field to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only These are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts should fall within the scope of protection of this application.

It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or equipment that includes a series of steps or units and need not be limited to clearly defined terms. Those steps or elements listed may instead include other steps or elements not expressly listed or inherent to the process, method, product or apparatus.

It should be noted that, as long as there is no conflict, the embodiments and features in the embodiments of this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

Bone fusion of the knee joint can generally be divided into the following types: Bone fusion in the knee flexion position. Bone fusion in extension position, simultaneous fusion of one hip and knee, simultaneous fusion of one hip, knee, and ankle, knee subluxation Subpositional bony fusion. The main causes of bony fusion of the knee joint include ankylosing spondylitis, rheumatoid arthritis, bone tuberculosis, and synovectomy. The current implementation of surgery for bony fusion knee cases relies on the experience of the surgeon and the assistance of an osteotomy guide to complete the osteotomy. The existing osteotomy guide design method relies on manual experience to manually mark the three-dimensional knee joint model to mark some parameters for designing the osteotomy guide, and manufacture the osteotomy guide based on the design parameters.

In view of the technical problems of related technologies, this application provides a design method for an intelligent positioning device for complex bony fusion knee joints, as detailed in the following embodiments.

Figure 1 schematically shows a flow chart of a design method for an intelligent positioning device for complex bony fusion knee joints according to an embodiment of the present application.

As shown in Figure 1, the intelligent positioning device design method for complex bony fusion knee joint may include steps 110 to 150.

In step 110, the medical image of the bony fusion knee joint is segmented through a three-dimensional reconstruction model, and a three-dimensional image of the bony fusion knee joint is generated based on the segmentation result.

In step 120, the medical image of the bony fusion knee joint is recognized and processed through a key point recognition model to obtain multiple key points of the bony fusion knee joint.

In step 130, the medical image of the bony fusion knee joint is recognized and processed by fitting a region recognition model to obtain two fitting parts of the bony fusion knee joint.

In step 140, a positioning device for the bony fusion knee joint is generated based on the multiple key points of the bony fusion knee joint and the two fitting parts of the bony fusion knee joint.

In step 150, multiple key points of the bony fusion knee joint, two fitting parts of the bony fusion knee joint, and the bony fusion knee joint are displayed on the three-dimensional image of the bony fusion knee joint. Fusion knee positioning device.

This method can identify and process the medical image of the bony fusion knee joint through the key point recognition model, obtain multiple key points of the bony fusion knee joint, and perform the medical image processing of the bony fusion knee joint by fitting the region recognition model. Through recognition processing, two fitting parts of the bony fusion knee joint are obtained, and based on multiple key points of the bony fusion knee joint and the two fitting parts of the bony fusion knee joint, a positioning device for the bony fusion knee joint is generated. , designing the positioning device in this way eliminates the need to rely on manual experience to manually mark parameters, which can improve the efficiency of making the positioning device. And improve the matching degree between the positioning device and the bony fusion knee joint, thereby improving the accuracy when using the positioning device for osteotomy.

In some embodiments of the present application, the above-mentioned medical images of the bony fusion knee joint may be obtained through digital scanning technology, such as using CT scanning technology to obtain CT scan images of relevant parts of the knee joint.

In some embodiments of the present application, segmenting the medical image of the bony fused knee joint through the above three-dimensional reconstruction model, and generating the three-dimensional image of the bony fused knee joint based on the segmentation results includes: converting the medical image of the bony fused knee joint into The image is input to the UNet model, and the medical image of the bony fusion knee joint is roughly segmented through the UNet model to obtain a coarse segmentation image. The coarse segmentation image is then input to the PonitRend model, and the coarse segmentation image is segmented at the pixel level through the PonitRend model. , generating a three-dimensional image of the bony fusion knee joint based on the results of pixel-level segmentation.

In some embodiments of the present application, the above-mentioned three-dimensional reconstruction model may include a UNet model and a PonitRend model. Among them, the UNet model can be used to roughly segment medical images of bony fused knee joints. The PonitRend model is used for pixel-level segmentation of coarsely segmented images.

Referring specifically to Figure 2, the above three-dimensional reconstruction model consists of the UNet+PointRend model, which is used to segment the bone area in the image. UNet is mainly composed of two parts: feature extraction and feature restoration. The feature extraction part consists of convolution and pooling operations. The convolution operation includes convolution, batch normalization, activation and other operations. The feature restoration part is mainly completed by convolution and upsampling operations. In this embodiment, the upsampling method may be a bilinear interpolation method, which is used to restore the size of the feature map. PointRend is mainly composed of convolution and fully connected layers. PointRend can reclassify the boundaries of rough segmentation results to obtain better boundary segmentation effects, and can more accurately identify bone edge areas to achieve more accurate recognition results. For example, the UNet model is used to identify the target location area in a medical image of a bony fusion knee joint (for example, the area where the tibia and femur are bonyly fused), and performs an analysis on the medical image of the bony fusion knee joint based on the boundaries of the target location area. Coarse segmentation processing. Point Rend is used to reclassify the boundaries of the rough segmentation results to obtain boundary segmentation results that meet the preset conditions, thereby obtaining better boundary segmentation effects. Based on the boundary segmentation results, a two-dimensional medical image of the target location area is output, and a three-dimensional reconstruction is performed based on the two-dimensional medical image of the target location area to obtain a three-dimensional image of the bony fusion knee joint.

In some embodiments of the present application, identifying and processing the medical image of the bony fusion knee joint through a key point recognition model to obtain multiple key points of the bony fusion knee joint includes: inputting the medical image of the bony fusion knee joint Go to the Hourglass model, and use the Hourglass model to detect the features in the medical image of the bony fused knee joint, and obtain the center point of the femoral head, the apex of the intercondylar notch, the center point of the ankle joint, the lowest point of the internal and external tibial plateau, and the femur of the bony fused knee joint. The most distal point of the condyle.

Referring to Figure 3, the above-mentioned Hourglass model can be an hourglass structure, used to identify and process medical images of bony fusion knee joints, and output pixel-level predictions. The hourglass structure can be composed of a convolution layer (C1-C7) and a pooling layer. The feature map (C1a-C4a) in the middle part can be a copy layer of the convolution layer. The copy layer and the corresponding layer in the convolution layer are upsampled. By adding, new feature information can be obtained to achieve the effect of feature fusion, which is the C1b-C4b part in the figure. The entire hourglass structure is symmetrical, so that for every network layer in the process of obtaining low-resolution features, there will be a corresponding network layer in the upsampling process. Then the feature layers are superimposed to obtain a large feature layer, namely C1b, which retains the information of all layers and is equal to the size of the input original image. Then a heatmap representing the probability of key points is generated through 1x1 convolution. The point with the maximum probability value in the heat map is taken as the feature point, and the location of the feature point is the predicted key point location. In this embodiment, the characteristic points may include, but are not limited to, the center point of the femoral head, the apex of the intercondylar notch, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the most distal point of the femoral condyle of the bony fusion knee joint.

In some embodiments of the present application, the medical image of the bony fusion knee joint is recognized and processed by fitting a region recognition model, and the two fitting parts of the bony fusion knee joint are obtained by: combining the medical images of the bony fusion knee joint. The image is input to the deeplabv3+ model, and the medical image of the bony fused knee joint is recognized and processed by the deeplabv3+ model, and the climbing area of the anterior femoral condyle and the upper medial area of the tibial tubercle of the bony fused knee joint are obtained.

Referring to Figure 4, the structure of the above deeplabv3+ model can include Encoder and Decoder. Encoder can contain a DCNN (deep convolutional network) and an ASPP network. DCNN is a backbone network used to extract medical image features of bony fusion knee joints. The ASPP network consists of a 1*1 convolution, three 3*3 atrous convolutions and a global pooling. It is used to process the output of the backbone network. The ASPP network processes the output results of the backbone network at different sampling rates. The atrous convolution parallel sampling can better capture the contextual information of the image, and then concatenate the results and use a 1*1 convolution to reduce the number of channels. Decoder can transform the intermediate output of the backbone network and the output of ASPP to obtain the same shape, then connect them and perform 3*3 Convolution, and finally use the convolution results to achieve segmentation. Decoder can be an upsampling process, that is, a feature restoration process, which restores the feature map to be consistent with the input image size. In this way, the above-mentioned deeplabv3+ model can be used to segment the anterior femoral condyle climbing area and the upper medial area of the tibial tubercle in medical images of bony fusion knee joints.

In some embodiments of the present application, based on multiple key points of the bony fusion knee joint and two fitting parts of the bony fusion knee joint, generating a positioning device for the bony fusion knee joint includes: according to the bony fusion knee joint The center point of the femoral head, the apex of the intercondylar fossa, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the most distal point of the femoral condyle, respectively determine the joint line dart groove, femoral dart groove, and tibial dart groove on the positioning device. position information; according to the femoral anterior condyle climbing area and the upper medial area of the tibial tubercle of the bony fusion knee joint, two fitting areas of the positioning device are fitted, and the two fitting areas include the femoral pseudo The fitting area and the tibial fitting area; based on the position information of the joint line dart groove, the femoral dart groove, and the tibial dart groove in the positioning device and the femoral fitting area and tibial fitting area of the positioning device, the positioning of the bony fusion knee joint is generated device.

In some embodiments of the present application, the joint line dart is determined based on the center point of the femoral head, the apex of the intercondylar notch, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the most distal point of the femoral condyle of the bony fused knee joint. The position information of the groove, the femoral dart groove, and the tibial dart groove on the positioning device includes: determining the position information of the joint line dart groove on the positioning device based on the lowest point of the internal and external tibial platform and the most distal point of the femoral condyle, and the joint line dart groove The number of the femoral dart groove is 1; according to the center point of the femoral head and the vertex of the intercondylar notch, the position information of the femoral dart groove in the positioning device is determined, and the number of the femoral dart groove is 2; according to the vertex of the intercondylar notch and the center point of the ankle joint, the tibial dart groove is determined Position information of the dart slots on the positioning device, and the number of the tibial dart slots is 2.

In some embodiments of the present application, the position information of the joint line dart groove in the positioning device is determined based on the lowest point of the internal and external tibial platform and the most distal point of the femoral condyle. For example, the femoral and tibial joint lines are determined based on the lowest point of the internal and external tibial platform and the most distal point of the femoral condyle, and then the position information of the joint line dart groove in the positioning device is determined based on the femoral and tibial joint lines.

In some embodiments of the present application, the position information of the femoral dart groove in the positioning device is determined based on the center point of the femoral head and the apex of the intercondylar notch. For example, the mechanical axis of the femur is determined based on the center point of the femoral head and the apex of the intercondylar notch, and then the position information of the femoral dart groove on the positioning device is determined based on the mechanical axis of the femur. In this embodiment, the position information of the femoral dart groove in the positioning device can be based on The position information of the joint line dart groove in the positioning device is adjusted, and the position information of the femoral dart groove in the positioning device is adjusted according to the femoral anatomical line.

In some embodiments of the present application, the position information of the tibial dart groove on the positioning device is determined based on the vertex of the intercondylar notch and the center point of the ankle joint. For example, the mechanical axis of the tibia is determined based on the apex of the intercondylar notch and the center point of the ankle joint, and then the position information of the tibial dart groove on the positioning device is determined based on the mechanical axis of the tibia. In this embodiment, the position information of the tibial dart groove on the positioning device can be adjusted according to the position information of the joint line dart groove on the positioning device, and the position information of the joint line dart groove on the positioning device can be adjusted according to the tibial anatomical line.

In some embodiments of the present application, the location of the nail holes of the positioning device is determined based on the identification of the material used for the positioning device and the medical surgical planning scheme.

In some embodiments of the present application, multiple key points of the bony fusion knee joint, two fitting locations of the bony fusion knee joint, and the three-dimensional image of the bony fusion knee joint are displayed. The positioning device includes: the femoral head in the three-dimensional image of the bony fusion knee joint displays the center point of the femoral head; the tibia in the three-dimensional image of the bony fusion knee joint displays the center point of the ankle joint; The junction of the femur and tibia shows the apex of the intercondylar notch, the lowest point of the medial and lateral tibial plateau, and the most distal point of the femoral condyle in The upper medial area displays the matching positioning device of the bony fusion knee joint.

The positioning device of the bony fusion knee joint designed through the above method can be displayed in the three-dimensional image of the bony fusion knee joint, refer to Figures 7 and 8. During the operation, screws are used to fix the positioning device at the position of the bony fusion knee joint through the nail holes 4. For example, during fixation, the climbing area of the anterior femoral condyle fits the fitting area 5 of the positioning device, and the upper and inner area of the tibial tubercle fits the fitting area 6 of the positioning device. The femoral dart groove 1 can be used to cut the femur, and the tibial dart groove 2 can be used to cut the tibia.

This application provides an intelligent positioning device for complex bony fusion knee joints. The positioning device includes an osteotomy body. The osteotomy body includes a femoral side connecting component, an osteotomy component, and a tibial side connecting component. The femoral side connecting component and the tibial side connecting component are respectively located on both sides of the osteotomy component; the osteotomy component is provided with two femoral dart grooves, two tibial dart grooves, and one joint line dart groove; the femoral dart groove, the tibial dart groove, and The position of the joint line dart groove in the osteotomy component is determined by using a key point recognition model to identify multiple key points obtained from medical images of bony fusion knee joints; the shape of the fitting area of the femoral side connecting component The area of the anterior femoral condyle climbing area obtained by identifying the medical image of the bony fusion knee joint through a fitting area recognition model; the shape of the fitting area of the tibial side connecting component is determined through the fitting area recognition model The upper medial area of the tibial tubercle obtained by identifying the medical image of the bony fusion knee joint is determined; the femoral side connecting component and the tibial side connecting component are respectively provided with nail holes, and the nail holes are determined according to the osteotomy The identification of the ontology and the medical surgical planning plan are determined.

Referring to Figures 5 and 6, the positioning device includes an osteotomy body. The osteotomy body includes a femoral side connecting component, an osteotomy component, and a tibial side connecting component. The femoral side connecting component and the tibial side connecting component are respectively located at both sides of the osteotomy component. The osteotomy component is provided with two femoral dart grooves 1, two tibial dart grooves 2, and one joint line dart groove 3. A fitting area 5 is provided on the femoral side connecting component, and the shape of the fitting area 5 may be M-shaped. A fitting area 6 is provided on the tibial side connecting component, and the shape of the fitting area 6 may be an irregular circle.

In this embodiment, the design parameters of the above-mentioned positioning device can be determined through the above-mentioned intelligent positioning device design method for complex bony fusion knee joints. For example, this design method can be used to determine the position of the femoral dart groove 1 and the tibial dart groove 2. position, the position and shape of the fitting area 5, the position and shape of the fitting area 6, etc. The specific implementation process can be based on the above design method and will not be described again here.

FIG9 schematically shows a block diagram of a system for designing an intelligent positioning device for a complex bone fusion knee joint according to an embodiment of the present application.

As shown in FIG. 9 , the intelligent positioning device design system 600 for complex bony fusion knee joint may include a three-dimensional reconstruction module 610 , a key point identification module 620 , a fitting part identification module 630 , a positioning device generation module 640 and a display module 650 .

The three-dimensional reconstruction module 610 is configured to perform segmentation processing on the medical image of the bony fusion knee joint through the three-dimensional reconstruction model, and generate a three-dimensional image of the bony fusion knee joint based on the segmentation results.

The key point recognition module 620 is configured to perform recognition processing on the medical image of the bony fusion knee joint through a key point recognition model to obtain multiple key points of the bony fusion knee joint.

The fitting part identification module 630 is configured to identify the bony features by fitting the region identification model. The medical image of the fused knee joint is recognized and processed to obtain two fitting parts of the bony fused knee joint.

The positioning device generating module 640 is configured to generate a positioning device of the bony fusion knee joint based on a plurality of key points of the bony fusion knee joint and two fitting parts of the bony fusion knee joint.

The display module 650 is configured to display a plurality of key points of the bony fusion knee joint, the two fitting parts of the bony fusion knee joint, and the three-dimensional image of the bony fusion knee joint. Positioning device for bony fused knees.

The complex bony fusion knee joint intelligent positioning device design system 600 can identify and process the medical image of the bony fusion knee joint through the key point recognition model, obtain multiple key points of the bony fusion knee joint, and through fitting The region recognition model recognizes and processes the medical image of the bony fusion knee joint, and obtains two fitting parts of the bony fusion knee joint, and based on the multiple key points of the bony fusion knee joint and the two fitting parts of the bony fusion knee joint. Fitting parts to generate a positioning device for the bony fusion knee joint. Designing the positioning device in this way does not need to rely on manual experience to manually mark parameters. This can improve the efficiency of making the positioning device and improve the accuracy of the positioning device and the bony fusion knee joint. matching, thus improving the accuracy of osteotomy using this positioning device.

In some embodiments of the present application, the intelligent positioning device design system 600 for complex bony fusion knee joints can be used to implement the intelligent positioning device design method for complex bony fusion knee joints described in the embodiment of FIG. 1 .

Optionally, the above-mentioned three-dimensional reconstruction module 610 may be configured to: input the medical image of the bony fusion knee joint into the UNet model, and perform rough segmentation on the medical image of the bony fusion knee joint through the UNet model, Obtain a rough segmentation image; input the rough segmentation image into the PonitRend model, perform pixel-level segmentation on the rough segmentation image through the PonitRend model, and generate a three-dimensional image of the bony fusion knee joint based on the pixel-level segmentation results.

Optionally, the above-mentioned key point identification module 620 may be configured to: input the medical image of the bony fusion knee joint into the Hourglass model, and detect the features in the medical image of the bony fusion knee joint through the Hourglass model. , to obtain the center point of the femoral head, the apex of the intercondylar fossa, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the most distal point of the femoral condyle of the bony fusion knee joint.

Optionally, the above-mentioned fitting part identification module 630 may be configured to: input the medical image of the bony fusion knee joint into the deeplabv3+ model, and identify the medical image of the bony fusion knee joint through the deeplabv3+ model. Processing is performed to obtain the climbing area of the anterior femoral condyle and the upper medial area of the tibial tubercle of the bony fusion knee joint.

Optionally, the above-mentioned positioning device generation module 640 may be configured to: based on the center point of the femoral head, the apex of the intercondylar notch, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the farthest point of the femoral condyle of the bony fusion knee joint The endpoints respectively determine the position information of the joint line dart groove, the femoral dart groove, and the tibial dart groove in the positioning device; according to the climbing area of the anterior femoral condyle of the bony fusion knee joint and the upper and inner side of the tibial tubercle area, fitting two fitting areas of the positioning device, the two fitting areas include a femoral fitting area and a tibial fitting area; based on the joint line dart groove, the femoral dart groove, and the The tibial dart groove generates the positioning device of the bony fusion knee joint based on the position information of the positioning device and the femoral fitting area and tibial fitting area of the positioning device.

Optionally, the above display module 650 may be configured to: display the femoral head center point in the three-dimensional image of the bony fusion knee joint; display the tibia in the three-dimensional image of the bony fusion knee joint. Display the center point of the ankle joint; display the apex of the intercondylar notch, the lowest point of the internal and external tibial platform, and the most distal end of the femoral condyle at the junction of the femur and tibia in the three-dimensional image of the bony fused knee joint points; and in the three-dimensional image of the bony fusion knee joint, the area on the slope of the anterior femoral condyle and the upper medial area of the tibial tubercle display the positioning device of the bony fusion knee joint that matches them.

Regarding the specific limitations on the design system of the intelligent positioning device 600 for complex bony fusion knee joints, please refer to the above limitations on the design method of the intelligent positioning device for complex bony fusion knee joints, which will not be described again here. Each module in the above-mentioned intelligent positioning device design system for complex bony fused knee joints can be realized in whole or in part through software, hardware and their combinations. Each of the above modules may be embedded in or independent of the processor of the computer device in the form of hardware, or may be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the steps in the above embodiments are implemented. The computer device may be a terminal, and its internal structure diagram may be as shown in Figure 10. The computer equipment includes a processor, memory, network interface, display screen and input device connected by a system bus. Wherein, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes non-volatile storage media and internal memory. The non-volatile storage medium stores operating systems and computer programs. This internal memory provides an environment for the execution of operating systems and computer programs in non-volatile storage media. The network interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by a processor, a method for designing an intelligent positioning device for complex bony fusion knee joints is implemented. The display screen of the computer device may be a liquid crystal display or an electronic ink display. The input device of the computer device may be a touch layer covered on the display screen, or may be a button, trackball or touch pad provided on the computer device shell. , it can also be an external keyboard, trackpad or mouse, etc.

Those skilled in the art can understand that the structure shown in Figure 10 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Specific computer equipment can May include more or fewer parts than shown, or combine certain parts, or have a different arrangement of parts.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above embodiments are implemented.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage. media, when the computer program is executed, It may include the processes of the embodiments of each of the above methods. Any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and/or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

The technical features of the above embodiments can be combined in any way. To simplify the description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, all possible combinations should be used. It is considered to be within the scope of this manual.

The above-described embodiments only express several implementation modes of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all fall within the protection scope of the present application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

An intelligent positioning device design method for complex bony fusion knee joints, including:

Segment the medical image of the bony fusion knee joint through a three-dimensional reconstruction model, and generate a three-dimensional image of the bony fusion knee joint based on the segmentation results;

The medical image of the bony fusion knee joint is recognized and processed through a key point recognition model to obtain multiple key points of the bony fusion knee joint;

Perform recognition processing on the medical image of the bony fusion knee joint by fitting a region recognition model to obtain two fitting parts of the bony fusion knee joint;

Generate a positioning device for the bony fusion knee joint based on the multiple key points of the bony fusion knee joint and the two fitting parts of the bony fusion knee joint;

The three-dimensional image of the bony fusion knee joint shows multiple key points of the bony fusion knee joint, two fitting parts of the bony fusion knee joint, and the positioning of the bony fusion knee joint. device.
The method according to claim 1, wherein said segmenting the medical image of the bony fusion knee joint through the three-dimensional reconstruction model, and generating the three-dimensional image of the bony fusion knee joint based on the segmentation result includes: :

Input the medical image of the bony fusion knee joint into the UNet model, perform rough segmentation on the medical image of the bony fusion knee joint through the UNet model, and obtain a rough segmentation image;

The coarse segmentation image is input to the PonitRend model, and the coarse segmentation image is segmented at the pixel level through the PonitRend model, and a three-dimensional image of the bony fusion knee joint is generated based on the result of the pixel level segmentation.
The method according to claim 1, wherein the medical image of the bony fusion knee joint is recognized and processed by the key point recognition model to obtain a plurality of key points of the bony fusion knee joint including:

The medical image of the bony fusion knee joint is input into the Hourglass model, and the features in the medical image of the bony fusion knee joint are detected through the Hourglass model to obtain the center point of the femoral head of the bony fusion knee joint, The apex of the intercondylar fossa, the center point of the ankle joint, the lowest point of the medial and lateral tibial plateau, and the most distal point of the femoral condyle.
The method according to claim 3, wherein the medical image of the bony fusion knee joint is recognized and processed by fitting a region recognition model to obtain both sides of the bony fusion knee joint. The fitting parts include:

The medical image of the bony fusion knee joint is input into the deeplabv3+ model, and the medical image of the bony fusion knee joint is recognized and processed by the deeplabv3+ model to obtain the anterior femoral condyle climbing of the bony fusion knee joint. area and the upper medial area of the tibial tubercle.
The method according to claim 4, wherein the positioning device of the bony fusion knee joint is generated based on a plurality of key points of the bony fusion knee joint and two fitting parts of the bony fusion knee joint. include:

According to the femoral head center point, the intercondylar notch apex, the ankle joint center point, the lowest point of the tibial medial and lateral platforms, and the most distal point of the femoral condyle of the bony fusion knee joint, the position information of the joint line dart slot, the femoral dart slot, and the tibial dart slot in the positioning device are determined respectively;

Two fitting areas of the positioning device are fitted according to the climbing area of the anterior femoral condyle and the upper medial area of the tibial tubercle of the bony fusion knee joint, and the two fitting areas include the femoral fitting area. and tibia fitting area;

The bony fusion is generated based on the position information of the joint line dart groove, the femoral dart groove, and the tibial dart groove on the positioning device and the femoral fitting area and tibial fitting area of the positioning device. Knee joint positioning device.
The method according to claim 5, wherein the center point of the femoral head, the apex of the intercondylar fossa, the center point of the ankle joint, the lowest point of the internal and external tibial platform, and the most distal point of the femoral condyle of the bonyly fused knee joint , respectively determining the position information of the joint line dart groove, femoral dart groove, and tibial dart groove on the positioning device includes:

Determine the position information of the joint line dart groove in the positioning device according to the lowest point of the inner and outer tibial platform and the most distal end point of the femoral condyle, and the number of the joint line dart groove is 1;

According to the center point of the femoral head and the apex of the intercondylar notch, the position information of the femoral dart groove in the positioning device is determined, and the number of the femoral dart groove is 2;

The position information of the tibial dart grooves on the positioning device is determined based on the vertex of the intercondylar notch and the center point of the ankle joint, and the number of the tibial dart grooves is 2.
The method according to claim 1, wherein the three-dimensional image of the bony fusion knee joint displays a plurality of key points of the bony fusion knee joint and two virtual images of the bony fusion knee joint. The joint part and the positioning device of the bony fusion knee joint include:

The femoral head in the three-dimensional image of the bony fused knee joint shows the femoral head center point;

The tibia in the three-dimensional image of the bony fused knee joint shows the ankle joint center point;

Demonstrating the apex of the intercondylar notch, the lowest point of the medial and lateral tibial plateau, and the most distal point of the femoral condyle at the junction of the femur and tibia in the three-dimensional image of the bony fused knee joint; and

In the three-dimensional image of the bony fusion knee joint, the area where the anterior femoral condyle climbs and the upper medial area of the tibial tubercle shows the matching positioning device of the bony fusion knee joint.
An intelligent positioning device design system for complex bony fusion knee joints, including:

The three-dimensional reconstruction module is configured to segment the medical image of the bony fusion knee joint through the three-dimensional reconstruction model, and generate a three-dimensional image of the bony fusion knee joint based on the segmentation results;

A key point identification module configured to identify and process the medical image of the bony fusion knee joint through a key point identification model to obtain multiple key points of the bony fusion knee joint;

The fitting part identification module is configured to perform identification processing on the medical image of the bony fusion knee joint through a fitting area recognition model, and obtain two fitting parts of the bony fusion knee joint;

a positioning device generation module configured to generate a positioning device for the bony fusion knee joint based on a plurality of key points of the bony fusion knee joint and two fitting parts of the bony fusion knee joint;

The display module is configured to display multiple key points of the bony fusion knee joint, two fitting parts of the bony fusion knee joint, and a positioning device of the bony fusion knee joint on the three-dimensional image of the bony fusion knee joint.
An intelligent positioning device for complex bony fusion knee joints. The positioning device includes an osteotomy body. The osteotomy body includes a femoral side connecting component, an osteotomy component, and a tibial side connecting component. The femoral side connecting component The connecting component and the tibial side connecting component are respectively located on both sides of the osteotomy component;

The osteotomy component is provided with two femoral dart grooves, two tibial dart grooves, and a joint line dart groove; the femoral dart groove, the tibial dart groove, and the joint line dart groove are located in the osteotomy The positions of the components are determined by using multiple key points obtained by identifying medical images of bony fused knee joints using a key point recognition model;

The shape of the fitting area of the femoral side connecting component is determined by identifying the climbing area of the femoral anterior condyle obtained from the medical image of the bony fusion knee joint by a fitting area recognition model;

The shape of the fitting area of the tibial side connecting component is determined by identifying the upper medial area of the tibial tubercle obtained by identifying the medical image of the bony fusion knee joint using the fitting area recognition model;

The femoral side connecting component and the tibial side connecting component are respectively provided with nail holes, and the nail holes are Determined based on the identification of the osteotomy body and the medical surgical planning plan.
A computer device, including a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the computer program, the method of any one of claims 1 to 7 is implemented. step.