A knee prosthesis design method, system, electronic device and product

Abstract

The present application belongs to the technical field of prosthesis manufacturing, and aims to provide a knee joint prosthesis design method, system, electronic device and product. The present application firstly constructs an initial prosthesis design model for a specified KBD knee osteoarthritis patient, then obtains knee joint image data of the specified KBD knee osteoarthritis patient, and obtains an image feature set according to the knee joint image data. Subsequently, multiple sets of the image feature set are divided into a training set, a validation set and a test set, and a final prosthesis design model is obtained based on the training set, the validation set and the test set. The final prosthesis design model is used to obtain a knee joint prosthesis structure model matched with the specified KBD knee osteoarthritis patient, and then a knee joint prosthesis matched with the specified KBD knee osteoarthritis patient is manufactured. Thus, the adaptation between the knee joint prosthesis and the skeleton of the specified KBD knee osteoarthritis patient can be improved, and the comfort of the patient after installing the knee joint prosthesis and the service life of the prosthesis can be improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of prosthesis manufacturing, and particularly relates to a knee joint prosthesis design method and system, an electronic device and a product. BACKGROUND

[0002] Knee joint is the largest joint in human body, with complex anatomy and high requirement for movement function. Kashin-Beck disease (KBD) is a kind of endemic and chronic deforming osteoarthropathy, which is an independent bone and joint disease. The disease originates from epiphyseal cartilage, epiphyseal plate cartilage and articular cartilage during childhood development, causing degeneration and deep cartilage cell necrosis. Since the 1960s and 1970s, the prevalence of the disease and the gradual development of prevention methods have been carried out, and currently there is no reported new case of adolescent disease, and those who have suffered from the disease have entered middle age and have shown obvious deforming osteoarthritis of the lower extremity joints (hips, knees and ankles), which seriously damages the health of residents in the disease area and is one of the important public health problems.

[0003] KBD-induced knee osteoarthritis (hereinafter referred to as "KBD knee arthritis") seriously affects the walking, labor, work and daily life of patients. According to the diagnosis and treatment scheme of orthopedics, patients with knee arthritis need to replace the joint to restore basic mobility. At present, there is no artificial knee joint prosthesis suitable for KBD knee arthritis at home and abroad, and the artificial knee joint used by patients with KBD knee arthritis during surgery is usually "borrowed" from the existing artificial knee joint prosthesis for senile degenerative osteoarthritis. However, in the process of using the prior art, the inventors found that at least the following problems exist in the prior art:

[0004] Since the age of onset and pathological mechanism of KBD knee arthritis are different from those of senile degenerative osteoarthritis, that is, KBD knee arthritis gradually occurs on the basis of developmental abnormalities of the distal femur and proximal tibia in childhood, while senile degenerative osteoarthritis occurs on the basis of normally developed joints, therefore, the two are different in etiology, pathology, clinic and imaging, and there are obvious differences in the anatomical parameters of the tibia and femur. The artificial knee joint prosthesis in the prior art often has problems such as poor compatibility with the bones of KBD knee arthritis patients and low bone coverage, which may lead to problems such as fast wear, unstable movement function and short service life after surgery.

[0005] Accordingly, the inventors collected a sufficient amount of knee joint data of KBD knee arthritis patients and printed them into 3D models, simulated the replacement of the existing knee joint prosthesis on the models, and obtained the following measurement results:

[0006] 1) For tibial plateau prosthesis, the ratio of anterior-posterior (AP) / medial-lateral (ML) of market conventional product is about 0.66, while the ratio of AP / ML of KBD knee osteoarthritis patients is about 0.72, so the AP size of patient's tibial plateau is larger than that of market conventional knee joint under the condition of the same ML size, and the ML size of patient's tibial plateau is smaller than that of market conventional knee joint under the condition of the same AP size;

[0007] 2) For femoral condyle prosthesis, the ratio of AP / ML of market conventional product is about 0.87-0.90, while the ratio of AP / ML of KBD knee osteoarthritis patients is about 0.86, so the AP size of patient's femoral condyle is smaller than that of market conventional knee joint under the condition of the same ML size, and the ML size of patient's femoral condyle is larger than that of market conventional knee joint under the condition of the same AP size.

[0008] The above results show that if the existing artificial knee joint prosthesis on the market is used, the sagittal plane bone resection amount of KBD knee osteoarthritis patients will be increased, and the small ML value will cause the coronal plane bone coverage rate to be reduced, thereby causing unstable movement function, severely reducing the patient's proprioception, and making the patient's gait unnatural during daily activities, especially when going up and down stairs, which affects their normal life. In addition, the mismatch between the prosthesis and the bone also affects the service life of the prosthesis, causing the service life of the prosthesis to be short.

[0009] Therefore, it is necessary to study a design method of a knee joint prosthesis suitable for patients with KBD knee osteoarthritis to make up for the clinical actual blank and provide the best treatment result for patients. SUMMARY

[0010] The present application aims to at least partially solve the above technical problems, and provides a knee joint prosthesis design method, system, electronic device and product.

[0011] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme:

[0012] In a first aspect, the present application provides a knee joint prosthesis design method, comprising:

[0013] constructing an initial prosthesis design model of a specified KBD knee osteoarthritis patient;

[0014] obtaining knee joint image data of the specified KBD knee osteoarthritis patient, and obtaining an image feature set according to the knee joint image data; wherein the knee joint image data and the image feature set are provided with multiple groups, and the multiple groups of image feature sets are one-to-one corresponding to the multiple groups of knee joint image data;

[0015] dividing a plurality of groups of the image feature sets into a training set, a validation set and a test set;

[0016] training the initial prosthesis design model using the training set to obtain an initial trained prosthesis design model;

[0017] adjusting hyperparameters of the initial trained prosthesis design model using the validation set to obtain an adjusted prosthesis design model;

[0018] evaluating the adjusted prosthesis design model using the test set, and obtaining a final prosthesis design model after the evaluation passes, so as to obtain a knee prosthesis structure model matched with the specified KBD knee osteoarthritis patient according to the final prosthesis design model, and then manufacturing a knee prosthesis matched with the specified KBD knee osteoarthritis patient based on the knee prosthesis structure model.

[0019] In one possible design, the knee prosthesis structure model includes a femoral condyle prosthesis structure model and a tibial plateau prosthesis structure model matched with the femoral condyle prosthesis structure model; and in the tibial plateau prosthesis structure model, a connecting part of a tibial plateau and a medullary cavity stem is provided with an anti-rotation island wing type structure, and an outer surface of the anti-rotation island wing type structure is designed with a plurality of vertical grooves.

[0020] In one possible design, the knee image data of the specified KBD knee osteoarthritis patient is obtained by CT scanning; correspondingly, after obtaining the knee image data of the specified KBD knee osteoarthritis patient, the method further includes:

[0021] converting the knee image data into color image data by using a pseudo-color processing method.

[0022] In one possible design, after obtaining the image feature set according to the knee image data, the method further includes:

[0023] performing denoising processing on the image features in the image feature set to obtain a denoised image feature set, so as to divide a plurality of groups of the denoised image feature set into a training set, a validation set and a test set.

[0024] In one possible design, the number of groups of the image feature sets in the training set is m, and m is a natural number greater than 1; correspondingly, training the initial prosthesis design model using the training set to obtain an initial trained prosthesis design model includes:

[0025] According to the m sets of image feature sets in the training set, m sets of knee joint anatomical parameter sets corresponding to the m sets of image feature sets are obtained; wherein any set of knee joint anatomical parameter sets includes femoral medial-lateral diameter, femoral anterior-posterior diameter, femoral lateral condyle anterior-posterior diameter, femoral medial condyle anterior-posterior diameter, femoral medial condyle width and / or femoral lateral condyle width;

[0026] The m sets of knee joint anatomical parameter sets are used to train the initial prosthesis design model to obtain an initial trained prosthesis design model.

[0027] In one possible design, according to any set of image feature sets in the training set, a corresponding knee joint anatomical parameter set is obtained, including:

[0028] A knee joint three-dimensional model is constructed according to the knee joint image data corresponding to any set of image feature sets in the training set;

[0029] Knee joint anatomical parameter measurement is performed on the knee joint three-dimensional model to obtain a knee joint anatomical parameter set matched with the current image feature set.

[0030] In a second aspect, the present application provides a knee joint prosthesis design system for implementing the knee joint prosthesis design method as described in any one of the above aspects; the knee joint prosthesis design system comprises:

[0031] An initial model construction module is configured to construct an initial prosthesis design model for a specified KBD knee osteoarthritis patient;

[0032] A patient data acquisition module is configured to acquire knee joint image data of the specified KBD knee osteoarthritis patient and obtain an image feature set according to the knee joint image data; and further configured to divide multiple sets of the image feature sets into a training set, a validation set and a test set; wherein the knee joint image data and the image feature set are each provided with multiple sets, and multiple sets of the image feature sets are each provided in one-to-one correspondence with multiple sets of the knee joint image data;

[0033] A model optimization module is in communication connection with the initial model construction module and the patient data acquisition module, respectively, and is configured to train the initial prosthesis design model using the training set to obtain an initial trained prosthesis design model; configured to adjust the hyperparameters of the initial trained prosthesis design model using the validation set to obtain an adjusted prosthesis design model; and further configured to evaluate the adjusted prosthesis design model using the test set, and obtain a final prosthesis design model after the evaluation passes, so as to obtain a knee joint prosthesis structure model matched with the specified KBD knee osteoarthritis patient according to the final prosthesis design model, and further manufacture a knee joint prosthesis matched with the specified KBD knee osteoarthritis patient based on the knee joint prosthesis structure model.

[0034] In a third aspect, the present application provides an electronic device comprising:

[0035] a memory for storing computer program instructions; and,

[0036] a processor for executing the computer program instructions to complete the operation of the knee prosthesis design method according to any one of the preceding aspects.

[0037] In a fourth aspect, the present application provides a computer program product comprising computer programs or instructions, which, when executed by a computer, implement the knee prosthesis design method according to any one of the preceding aspects.

[0038] The present application has the following beneficial effects:

[0039] The present application discloses a knee prosthesis design method, system, electronic device and product, which can improve the matching degree between the knee prosthesis and the bones of the specified KBD knee osteoarthritis patient, improve the comfort of the patient after installing the knee prosthesis, and prolong the service life of the knee prosthesis. Specifically, in the implementation process of the present application, an initial prosthesis design model of the specified KBD knee osteoarthritis patient is first constructed, then knee joint image data of the specified KBD knee osteoarthritis patient is obtained, and a set of image features is obtained according to the knee joint image data. Then, the multiple sets of image feature sets are divided into a training set, a validation set and a test set, and a final prosthesis design model is obtained based on the training set, the validation set and the test set, so as to obtain a knee prosthesis structure model matched with the specified KBD knee osteoarthritis patient according to the final prosthesis design model, and then a knee prosthesis matched with the specified KBD knee osteoarthritis patient is manufactured based on the knee prosthesis structure model. In this process, the initial prosthesis design model is optimized by the knee joint image data of the specified KBD knee osteoarthritis patient, which can improve the matching degree and bone coverage rate between the finally obtained knee prosthesis and the patient's bones, reduce the joint wear of the patient, and thus can minimize the stimulation of the patient's knee joint soft tissue caused by the mismatch of the prosthesis, improve the postoperative patient satisfaction, the most natural motion function and gait, etc., and effectively prolong the service life of the prosthesis.

[0040] Other beneficial effects of the present application will be further described in the specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 is a flowchart of a knee prosthesis design method in Example 1;

[0042] Figure 2 is a schematic diagram of converting the example knee joint image data into color image data in Example 1;

[0043] Figure 3 is a schematic view of the installation of a femoral condyle prosthesis of an example existing knee prosthesis to a patient's femur;

[0044] Figure 4 is a schematic view of the installation of a femoral condyle prosthesis of a knee prosthesis obtained based on example 1 to a patient's femur;

[0045] Figure 5 is a schematic view of the installation of a tibial plateau prosthesis of an example existing knee prosthesis to a patient's tibia;

[0046] Figure 6 is a schematic view of the installation of a tibial plateau prosthesis of a knee prosthesis obtained based on example 1 to a patient's tibia;

[0047] Figure 7 is a block diagram of a module of a knee prosthesis design system in example 2;

[0048] Figure 8 is a block diagram of a module of an electronic device in example 3. DETAILED DESCRIPTION

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the present application will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the drawings is only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings. It should be noted that the description of these embodiments is used to help understand the present application, but does not constitute a limitation on the present application.

[0050] Example 1:

[0051] The present embodiment discloses a knee prosthesis design method, which can be executed by a computer device or a virtual machine with certain computing resources, such as a personal computer, a smart phone, a personal digital assistant, a wearable device, or the like, or by a virtual machine.

[0052] As shown in Figure 1 , a knee prosthesis design method can include, but is not limited to, the following steps:

[0053] S1. Construct an initial prosthesis design model for a specified KBD knee osteoarthritis patient;

[0054] S2. Obtain knee joint image data of the specified KBD knee osteoarthritis patient, and obtain an image feature set according to the knee joint image data; wherein the knee joint image data and the image feature set are each provided with multiple groups, and multiple groups of the image feature set are respectively and correspondingly provided with multiple groups of the knee joint image data; it should be understood that in the embodiment, any image feature in the image feature set is bound with a feature label.

[0055] In step S2, the knee joint image data of the specified KBD knee osteoarthritis patient is obtained by CT scanning; correspondingly, after obtaining the knee joint image data of the specified KBD knee osteoarthritis patient in step S2, the method further comprises:

[0056] The knee joint image data is converted into color image data by using a pseudo-color processing method. Figure 2 A schematic diagram for converting the example knee joint image data (Fig. (a)) into color image data (Fig. (b)).

[0057] Specifically, the pseudo-color (also known as "B-color") processing method is an image processing method for converting black and white image display into color display, which can be used in all gray-scale displayed ultrasound images in principle. In the embodiment, in the process of converting the knee joint image data into color image data by using the pseudo-color processing method, the image amplitude (gray-scale display) of the knee joint image data is first divided into many color domains, and then the gray-scale display is converted into color display by using the pseudo-color coding method, so that the black and white image data becomes color image data. Since the human eye is not very sensitive to the gray-scale level, the sensitivity of the human eye to different image feature intensities can be enhanced after the black and white image is converted into color image, the dynamic range of the display signal is increased subjectively, and thus the recognizable degree of the image boundary of the knee joint image data can be enhanced.

[0058] In the embodiment, after obtaining the knee joint image data of the specified KBD knee osteoarthritis patient, the knee joint image data is further converted into color image data by using the pseudo-color processing method, which can solve the problem that the lesions and the properties of some surfaces of ordinary images are not easy to identify, and can facilitate to improve the accuracy of image feature extraction in the image feature set obtained subsequently and the relationship between the image features and the feature labels, more accurately establish the relationship between the image features and the feature labels, and also facilitate the user to further directly observe and diagnose the lesion condition of the specified KBD knee osteoarthritis patient by using the color image data with the naked eye.

[0059] In step S2, after obtaining the image feature set according to the knee joint image data, the method further comprises:

[0060] The image features in the image feature set are denoised to obtain a denoised image feature set, so as to divide multiple groups of the denoised image feature set into a training set, a verification set and a test set.

[0061] It should be noted that the image noise in the image feature set comes from the internal texture and the surrounding background noise affecting the knee joint image data, and the image shape in different knee joint image data also has subtle changes. In the embodiment, the frequently occurring common features in multiple groups of image feature sets can be recognized through iterative training, so as to effectively eliminate the noise in the training sample, and at the same time, the subtle differences between different denoised image feature sets can be retained, so that the main shape features in the knee joint image data are learned.

[0062] S3. Divide multiple groups of the image feature set into a training set, a verification set and a test set; in the embodiment, multiple groups of the image feature set are randomly divided into a training set, a verification set and a test set. As an example, in the embodiment, the group number ratio of the image feature sets corresponding to the training set, the verification set and the test set can be set to 6:2:2, which is not limited here.

[0063] S4. Train the initial prosthesis design model using the training set to obtain an initial trained prosthesis design model.

[0064] The number of groups of the image feature set in the training set is m, and m is a natural number greater than 1. Correspondingly, in step S4, the initial prosthesis design model is trained using the training set to obtain an initial trained prosthesis design model, which includes:

[0065] S401. Obtain m groups of knee joint anatomical parameter sets corresponding to the m groups of image feature sets according to the m groups of image feature sets in the training set; wherein any group of knee joint anatomical parameter set includes medial-lateral (ML), anterior-posterior (AP), lateral anteroposterior (LAP), medial anteroposterior (MAP), medial condylar width (MCW) and / or lateral condylar width (LCW); it should be noted that the image feature set includes a plurality of knee joint anatomical parameters of KBD suitable for total knee arthroplasty (TKA), such as the above-mentioned medial-lateral, lateral anteroposterior, medial anteroposterior, medial condylar width and lateral condylar width, etc. which are not limited here.

[0066] In step S401, according to any one set of image feature sets in the training set, a set of knee joint anatomical parameters corresponding thereto is obtained, including:

[0067] S4011. A knee joint three-dimensional model is constructed according to the knee joint image data corresponding to any one set of image feature sets in the training set;

[0068] S4012. Knee joint anatomical parameter measurement is performed on the knee joint three-dimensional model to obtain a set of knee joint anatomical parameters matched with the current image feature set. Specifically, in the embodiment, Mimics (Materiaise's interactive medical image control system) software (a 3D image generation and editing processing software) is used to construct a knee joint three-dimensional model corresponding to the knee joint image data, so as to quickly obtain a set of knee joint anatomical parameters and perform subsequent model training.

[0069] S402. The initial prosthesis design model is trained using m sets of knee joint anatomical parameter sets to obtain an initial trained prosthesis design model.

[0070] S5. The initial trained prosthesis design model is adjusted in super parameters using the verification set to obtain an adjusted prosthesis design model.

[0071] S6. The adjusted prosthesis design model is evaluated using the test set, and a final prosthesis design model is obtained after the evaluation passes, so as to obtain a knee joint prosthesis structure model matched with the specified KBD knee osteoarthritis patient according to the final prosthesis design model. For example, the latest knee joint image feature data of the specified KBD knee osteoarthritis patient is obtained, and then input into the final prosthesis design model to obtain a knee joint prosthesis structure model matched with the specified KBD knee osteoarthritis patient. Then, the knee joint prosthesis matched with the specified KBD knee osteoarthritis patient can be manufactured based on the knee joint prosthesis structure model. In the embodiment, the knee joint prosthesis manufactured according to the final prosthesis design model is manufactured according to the actual situation of the knee joint of the specified KBD knee osteoarthritis patient, and has the highest matching degree with the specified KBD knee osteoarthritis patient.

[0072] It should be noted that in the embodiment, the training set is used to train the initial prosthesis design model, and after the model is trained, the verification set is used to evaluate the model effect, and at the same time, the model is adjusted in super parameters to be in the best state. Then, the model is evaluated by the test set. At this time, evaluation indexes such as accuracy, precision and recall rate can be obtained to judge whether the model passes the evaluation, and a final prosthesis design model is obtained after the evaluation passes.

[0073] Specifically, in this embodiment, the knee joint prosthesis structure model includes a femoral condyle prosthesis structure model and a tibial plateau prosthesis structure model matched with the femoral condyle prosthesis structure model; in the tibial plateau prosthesis structure model, the connecting part of the tibial plateau and the medullary cavity stem is provided with an anti-rotation island wing structure, and the outer surface of the anti-rotation island wing structure is designed with a plurality of vertical grooves. It should be noted that in this embodiment, the design of the anti-rotation island wing structure and the plurality of vertical grooves on the outer surface thereof can increase the contact area between the tibial plateau prosthesis in the prosthesis and the tibial bone of the specified KBD knee osteoarthritis patient, prevent loosening of the knee joint prosthesis, improve the stability of the knee joint prosthesis, and prolong the service life of the knee joint prosthesis.

[0074] It should also be noted that in this embodiment, the knee joint prosthesis matched with the specified KBD knee osteoarthritis patient is manufactured according to the final prosthesis design model, which is beneficial to optimize the AP / ML ratio of the femoral condyle prosthesis in the knee joint prosthesis, improve the coverage of the knee joint prosthesis and the bone, be closer to the natural shape of the femoral condyle, better simulate and restore the biomechanical properties of the knee joint of the patient, reduce stress concentration on the joint surface, reduce the risk of loosening and wear of the knee joint prosthesis, and thus prolong the service life of the knee joint prosthesis; in this embodiment, the design of the knee joint prosthesis is customized according to the condition of the knee joint of the patient, taking into account the specific condition of the patient, the morphology of the bone and the degree of joint damage, which can promote the prosthesis to simulate the movement trajectory of the natural knee joint after implantation, reduce the problem of joint wear and pain of the patient after surgery. In addition, in this embodiment, the knee joint prosthesis can be made by additive manufacturing process, and doctors can customize the treatment according to the individual differences of the patient to achieve the best effect, so as to further ensure that the knee joint prosthesis matches the bone structure of the patient, improve the surgical effect and patient comfort.

[0075] As an example, Figure 3 a schematic diagram of the installation of the femoral condyle prosthesis in the existing knee joint prosthesis and the patient's femur is given, Figure 3 Fig. (a) is a side view of the installation of a femoral condyle prosthesis and a patient's femur, in which the ML size of the femoral condyle prosthesis matches the patient's femur, and the AP size of the femoral condyle prosthesis is about 3 mm larger than the AP size of the patient's femur. The difference between the AP size of the femoral condyle prosthesis and the AP size of the patient's femur is the size difference d in the figure. Fig. (b) is a top view of the installation of another femoral condyle prosthesis and a patient's femur, in which the AP size of the femoral condyle prosthesis matches the AP size of the patient's femur, and the posterior condyle of the femoral condyle prosthesis is 3 mm smaller than the AP size of the patient's femur; Figure 4 a schematic diagram of the installation of the femoral condyle prosthesis in the knee joint prosthesis obtained based on this embodiment and the patient's femur is given, and the ML size and the AP size of the femoral condyle prosthesis match the patient's femur.Figure 5 Figures (a) and (b) are schematic diagrams of installation of two kinds of tibial platform prostheses in existing knee joint prostheses on patient tibia, wherein the AP size and the ML size of the two kinds of tibial platform prostheses cannot well meet the coverage of the patient's bone; Figure 6 is a schematic diagram of installation of a tibial platform prosthesis in a knee joint prosthesis obtained based on the present embodiment on patient tibia, and the tibial platform prosthesis obtained based on the present embodiment has higher matching degree with patient tibia.

[0076] The present embodiment can improve the matching degree between the knee joint prosthesis and the bone of the specified KBD knee osteoarthritis patient, thereby improving the comfort of the patient after installing the knee joint prosthesis, and prolonging the service life of the knee joint prosthesis. Specifically, in the implementation process of the present embodiment, an initial prosthesis design model of the specified KBD knee osteoarthritis patient is first constructed, then knee joint image data of the specified KBD knee osteoarthritis patient is obtained, and an image feature set is obtained based on the knee joint image data, then a plurality of sets of the image feature set are divided into a training set, a validation set and a test set, and a final prosthesis design model is obtained based on the training set, the validation set and the test set, so as to obtain a knee joint prosthesis structure model matched with the specified KBD knee osteoarthritis patient based on the final prosthesis design model, and then a knee joint prosthesis matched with the specified KBD knee osteoarthritis patient is manufactured based on the knee joint prosthesis structure model. In this process, the present embodiment optimizes the initial prosthesis design model through the knee joint image data of the specified KBD knee osteoarthritis patient, which can improve the matching degree and bone coverage between the finally obtained knee joint prosthesis and the patient's bone, thereby reducing the joint wear of the patient, and further minimizing the stimulation of the patient's knee joint soft tissue caused by the mismatch of the prosthesis, thereby improving the postoperative patient satisfaction, the most natural motion function and gait, etc., and effectively prolonging the service life of the prosthesis.

[0077] Embodiment 2:

[0078] The present embodiment discloses a knee joint prosthesis design system for implementing the knee joint prosthesis design method in Embodiment 1; as shown in Figure 7 The knee joint prosthesis design system comprises:

[0079] An initial model construction module for constructing an initial prosthesis design model of a specified KBD knee osteoarthritis patient;

[0080] A patient data acquisition module for acquiring knee joint image data of the specified KBD knee osteoarthritis patient, and obtaining an image feature set based on the knee joint image data; and for dividing a plurality of sets of the image feature set into a training set, a validation set and a test set; wherein the knee joint image data and the image feature set are each provided with a plurality of sets, and a plurality of sets of the image feature set are respectively and correspondingly provided with a plurality of sets of the knee joint image data.

[0081] a model optimization module, communicatively connected with the initial model construction module and the patient data acquisition module respectively, configured to train the initial prosthesis design model using the training set to obtain an initial trained prosthesis design model, to adjust hyperparameters of the initial trained prosthesis design model using the validation set to obtain an adjusted prosthesis design model, and to evaluate the adjusted prosthesis design model using the test set and to obtain a final prosthesis design model after the evaluation passes, so as to obtain a knee prosthesis structure model matched with the specified KBD knee osteoarthritis patient according to the final prosthesis design model, and to further manufacture a knee prosthesis matched with the specified KBD knee osteoarthritis patient based on the knee prosthesis structure model.

[0082] It should be noted that the working process, working details and technical effects of the knee prosthesis design system provided in Embodiment 2 can be referred to Embodiment 1, and will not be repeated here.

[0083] Embodiment 3

[0084] On the basis of Embodiment 1 or 2, the present embodiment discloses an electronic device, which can be a smartphone, a tablet computer, a notebook computer or a desktop computer, etc. The electronic device can be referred to as a user terminal, a portable terminal, a desktop terminal, etc., as shown in Figure 8 The electronic device includes:

[0085] a memory for storing computer program instructions; and

[0086] a processor for executing the computer program instructions to complete the operations of the knee prosthesis design method according to any one of Embodiments 1.

[0087] Specifically, the processor 301 can include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 301 can be implemented in at least one of a hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), a PLA (Programmable Logic Array). The processor 301 can also include a main processor and a coprocessor, the main processor being a processor for processing data in an awake state, also known as a CPU (Central Processing Unit), and the coprocessor being a low-power processor for processing data in a standby state. In some embodiments, the processor 301 can be integrated with a GPU (Graphics Processing Unit) for rendering and drawing content required to be displayed by the display screen.

[0088] The memory 302 can include one or more computer-readable storage media, which can be non-transitory. The memory 302 can also include a high-speed random access memory, and a nonvolatile memory such as one or more disk storage devices, flash storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 302 is used to store at least one instruction for being executed by the processor 301 to implement the knee prosthesis design method provided in Embodiment 1 of the present application.

[0089] In some embodiments, the terminal can also optionally include a communication interface 303 and at least one peripheral device. The processor 301, the memory 302 and the communication interface 303 can be connected through a bus or a signal line. Each peripheral device can be connected to the communication interface 303 through a bus, a signal line or a circuit board. Specifically, the peripheral device includes at least one of a radio frequency circuit 304, a display screen 305 and a power supply 306.

[0090] The communication interface 303 can be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 301 and the memory 302. In some embodiments, the processor 301, the memory 302 and the communication interface 303 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 301, the memory 302 and the communication interface 303 can be implemented on a separate chip or circuit board, and the present embodiment does not limit this.

[0091] The RF circuit 304 is used for receiving and transmitting RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 304 communicates with communication networks and other communication devices through electromagnetic signals.

[0092] The display screen 305 is used for displaying a UI (User Interface). The UI can include graphics, text, icons, video, and any combination thereof.

[0093] The power supply 306 is used for supplying power to various components in the electronic device.

[0094] Embodiment 4

[0095] On the basis of any one of the embodiments 1 to 3, the present embodiment discloses a computer program product comprising computer programs or instructions which, when executed by a computer, implement the knee prosthesis design method as described in any one of the embodiments 1.

[0096] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be realized by general computing devices, which can be concentrated on a single computing device or distributed on a network composed of multiple computing devices, and optionally, they can be realized by program codes executable by computing devices, so that they can be stored in storage devices and executed by computing devices, or they can be respectively manufactured into individual integrated circuit modules, or multiple modules or steps among them can be manufactured into a single integrated circuit module to realize. Thus, the present application is not limited to any specific combination of hardware and software.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of designing a knee prosthesis, characterized in that: The method comprises the following steps: constructing an initial prosthesis design model for a specified KBD knee osteoarthritis patient; obtaining knee joint image data of the specified KBD knee osteoarthritis patient, and obtaining an image feature set according to the knee joint image data; wherein the knee joint image data and the image feature set are provided with multiple groups, and multiple groups of the image feature set are provided in one-to-one correspondence with multiple groups of the knee joint image data; dividing the multiple groups of the image feature set into a training set, a validation set and a test set; training the initial prosthesis design model using the training set to obtain an initial trained prosthesis design model; adjusting the hyperparameters of the initial trained prosthesis design model using the validation set to obtain an adjusted prosthesis design model; evaluating the adjusted prosthesis design model using the test set, and obtaining a final prosthesis design model after the evaluation passes, so as to obtain a knee joint prosthesis structure model matched with the specified KBD knee osteoarthritis patient according to the final prosthesis design model, and then manufacturing a knee joint prosthesis matched with the specified KBD knee osteoarthritis patient based on the knee joint prosthesis structure model; the knee joint prosthesis structure model comprises a femoral condyle prosthesis structure model and a tibial plateau prosthesis structure model matched with the femoral condyle prosthesis structure model; wherein the tibial plateau prosthesis structure model is provided with an anti-rotation island wing structure at the connection part between the tibial plateau and the medullary cavity stem, and the outer surface of the anti-rotation island wing structure is designed with a plurality of vertical grooves; the number of groups of image feature sets in the training set is m, and m is a natural number greater than 1; correspondingly, the initial prosthesis design model is trained using the training set to obtain an initial trained prosthesis design model, which comprises: obtaining m groups of knee joint anatomical parameter sets corresponding to the m groups of image feature sets in the training set; wherein any group of knee joint anatomical parameter set comprises femoral medial-lateral diameter, femoral anterior-posterior diameter, femoral lateral condyle anterior-posterior diameter, femoral medial condyle anterior-posterior diameter, femoral medial condyle width and / or femoral lateral condyle width; training the initial prosthesis design model using the m groups of knee joint anatomical parameter sets to obtain an initial trained prosthesis design model.

2. The method of designing a knee prosthesis according to claim 1, wherein: The knee joint image data of the specified KBD knee osteoarthritis patient is obtained by CT scanning; correspondingly, after obtaining the knee joint image data of the specified KBD knee osteoarthritis patient, the method further comprises: using a pseudo-color processing method to convert the knee joint image data into color image data.

3. The method of designing a knee prosthesis according to claim 1, wherein: After obtaining the image feature set according to the knee joint image data, the method further comprises: performing denoising processing on the image features in the image feature set to obtain a denoised image feature set, so as to divide the multiple groups of denoised image feature sets into a training set, a validation set and a test set.

4. The method of designing a knee prosthesis of claim 1, wherein: According to any group of image feature set in the training set, a corresponding knee joint anatomical parameter set is obtained, which comprises: constructing a knee joint three-dimensional model according to the knee joint image data corresponding to any group of image feature set in the training set; The knee joint three-dimensional model is subjected to knee joint anatomical parameter measurement to obtain a knee joint anatomical parameter set matched with the current image feature set.

5. A knee prosthesis design system characterized by: The knee joint prosthesis design system comprises: An initial model construction module for constructing an initial prosthesis design model of a specified KBD knee osteoarthritis patient; A patient data acquisition module for acquiring knee joint image data of the specified KBD knee osteoarthritis patient and obtaining an image feature set according to the knee joint image data; and for dividing a plurality of sets of the image feature set into a training set, a verification set and a test set; wherein the knee joint image data and the image feature set are each provided with a plurality of sets, and the plurality of sets of the image feature set are each provided in one-to-one correspondence with the plurality of sets of the knee joint image data; A model optimization module in communication connection with the initial model construction module and the patient data acquisition module, for training the initial prosthesis design model using the training set to obtain an initial trained prosthesis design model; for adjusting the hyperparameters of the initial trained prosthesis design model using the verification set to obtain an adjusted prosthesis design model; and for evaluating the adjusted prosthesis design model using the test set and obtaining a final prosthesis design model after the evaluation passes, so as to obtain a knee joint prosthesis structure model matched with the specified KBD knee osteoarthritis patient according to the final prosthesis design model, and further manufacture a knee joint prosthesis matched with the specified KBD knee osteoarthritis patient based on the knee joint prosthesis structure model.

6. An electronic device, comprising: comprise: a memory for storing computer program instructions; and a processor for executing the computer program instructions to complete the operations of the knee joint prosthesis design method of any one of claims 1 to 4.

7. A computer program product comprising computer programs or instructions, characterized in that: The computer program or the instructions realize the knee joint prosthesis design method of any one of claims 1 to 4 when executed by a computer.

Images