Multi-disease artificial intelligence preoperative planning system for whole orthopedics department

Abstract

The invention provides a multi-disease artificial intelligence preoperative planning system for the whole orthopedics department, and the system comprises an importing module which is used for importing medical image data; the three-dimensional reconstruction module is used for establishing a corresponding three-dimensional image model according to the medical image data; the skeleton segmentation module is used for segmenting skeletons in the medical image data; and at least three of the following modules: a hip joint simulation module; a knee joint simulation module; a spine joint simulation module; a sports medicine simulation module; and a trauma operation simulation module. According to the pre-operation planning system, the multiple operation simulation modules are arranged and correspond to different types of orthopedic operations, and therefore the coverage range of the pre-operation planning system is greatly enlarged.

Description

Technical Field

[0001] The present application relates to the field of medical device technology, and more specifically, to an artificial intelligence preoperative planning system for multiple orthopedic diseases. Background Art

[0002] Traditional orthopedic surgery is planned independently by doctors, and the planning is overly dependent on the doctors’ experience, which may increase the risk of surgery and the accuracy and stability of the surgery are not high.

[0003] Based on this, a preoperative planning system using computer software can assist doctors in preoperative planning, but the current preoperative planning system is only targeted at one category of orthopedic surgery and has a relatively small scope of application. Summary of the invention

[0004] The problem addressed by this application is that existing preoperative planning systems are only targeted at one category of orthopedic surgery.

[0005] To solve the above problems, the present application provides an artificial intelligence preoperative planning system for all orthopedic diseases, which includes:

[0006] Import module, used to import medical imaging data;

[0007] A three-dimensional reconstruction module is used to establish a corresponding three-dimensional image model based on medical imaging data;

[0008] The bone segmentation module is used to segment bones in medical imaging data;

[0009] And at least three of the following modules:

[0010] Hip joint simulation module, used to simulate hip joint surgery process and determine the preoperative planning plan of hip joint surgery based on medical imaging data of hip joint;

[0011] Knee joint simulation module, used to simulate the knee joint surgery process and determine the preoperative planning plan of knee joint surgery based on the medical imaging data of the knee joint;

[0012] The spinal joint simulation module is used to simulate the spinal joint surgery process and determine the preoperative planning scheme of the spinal joint surgery based on the medical imaging data of the spinal joint;

[0013] Sports medicine simulation module, used to simulate the movement process of the knee joint based on the medical imaging data of the knee joint;

[0014] The trauma surgery simulation module is used to simulate the trauma surgery process and determine the preoperative planning plan for trauma surgery based on the medical imaging data of the trauma site.

[0015] Further, the hip joint simulation module includes a first key point recognition submodule, a first template measurement submodule, a first display submodule and a first guide plate submodule;

[0016] The first key point identification submodule is used to identify key points of the femur and acetabulum;

[0017] The first module measurement submodule is used to measure or calculate the anteversion angle, abduction angle, coverage rate of the acetabular cup prosthesis template, the difference in leg length and offset after surgery, and the amount and position of osteotomy at the proximal femur;

[0018] The first display submodule is used to display an image of the prosthesis template fitting with the bone, and to adjust it according to the operator's operation;

[0019] The first guide plate submodule is used to output the planned osteotomy amount and osteotomy position, and to derive an osteotomy guide plate model.

[0020] Furthermore, the key points identified by the first key point identification submodule include: the center of the left femoral head, the center of the femoral diameter, the top of the lesser trochanter, the lesser trochanter, the proximal point of the femur, the distal point of the femur, the medial epicondyle and the lateral epicondyle, the center of the right femoral head, the top of the lesser trochanter, the lesser trochanter, the proximal point of the femur, and the distal point of the femur.

[0021] Further, the knee joint simulation module includes a second key point recognition submodule, a second template measurement submodule, a second display submodule and a second guide plate submodule;

[0022] The second key point identification submodule is used to identify key points on the femur and tibia;

[0023] The second module measurement submodule is used to measure or calculate the amount of distal femoral osteotomy, the amount of posterior femoral condyle osteotomy, and the amount of tibial plateau osteotomy, and to calculate the sum of the amount of osteotomy on the medial side of the distal femur and the medial side of the tibial plateau, the sum of the amount of osteotomy on the lateral side of the distal femur and the lateral side of the tibial plateau, the sum of the amount of osteotomy on the medial side of the posterior femoral condyle and the medial side of the tibial plateau, and the sum of the amount of osteotomy on the lateral side of the posterior femoral condyle and the lateral side of the tibial plateau;

[0024] The second display submodule is used to display an image of the prosthesis template fitting with the bone, and to adjust it according to the operator's operation;

[0025] The second guide plate submodule is used to output the planned osteotomy amount and osteotomy position, and to export the osteotomy guide plate model.

[0026] Furthermore, the key points identified by the second key point identification submodule include: medial epicondyle, lateral epicondyle, distal medial side, distal lateral side, medial side of posterior condyle, lateral side of posterior condyle, anterior condyle, proximal point of femur and distal point of femur, medial point of tibial plateau, lateral point of plateau, medial 1 / 3 of tibial tuberosity, attachment point of posterior cruciate ligament, proximal point of tibia and distal point of tibia.

[0027] Further, the spinal joint simulation module includes a third key point recognition submodule, a third template measurement submodule, a third display submodule and a third guide plate submodule;

[0028] The third key point identification submodule is used to identify the key points of each vertebra of the spine;

[0029] The third module measurement submodule is used to measure or calculate the size and setting position of the screw module, the fusion device template, the titanium rod template and the cross-connection template;

[0030] The third display submodule is used to display the image of the prosthesis template fitting with the bone, and to adjust it according to the operation of the operator;

[0031] The third guide plate submodule is used to output the planned screw module, fusion cage template, titanium rod template and cross-connection template positions, and export the spinal guide plate model.

[0032] Further, the sports medicine simulation module includes a fourth key point recognition submodule, a fourth template measurement submodule and a fourth display submodule;

[0033] The fourth key point identification submodule is used to identify key points of the femur and tibia;

[0034] The fourth module measurement submodule is used to measure or calculate the length and expansion and contraction change of the bone tunnel at different flexion angles of the knee joint;

[0035] The fourth display submodule is used to display the dynamic effects of the knee joint at different flexion angles and the effects of bone tunnel changes.

[0036] Further, the trauma surgery simulation module includes a fifth key point recognition submodule, a fifth template measurement submodule, a fifth display submodule and a fifth guide plate submodule;

[0037] The fifth key point identification submodule is used to identify key points of the femur and acetabulum;

[0038] The fifth module measurement submodule is used to measure or calculate the size and setting position of the main nail module, the tail cap template, the spiral blade template and the interlocking screw template;

[0039] The fifth display submodule is used to display the image of the prosthesis template fitting with the bone, and to adjust it according to the operation of the operator;

[0040] The fifth guide plate submodule is used to output the planned positions of the main nail module, tail cap template, spiral blade template and interlocking screw template.

[0041] Furthermore, the knee joint simulation module also includes a unicompartmental submodule, which is connected to a second key point identification submodule, a second template measurement submodule, a second display submodule and a second guide plate submodule. When the unicompartmental submodule is in an activated state, the second key point identification submodule, the second template measurement submodule, the second display submodule and the second guide plate submodule only process operations corresponding to a single knee joint.

[0042] Furthermore, the bone segmentation module is provided with a pre-trained segmentation model, and the segmentation of the input medical image data includes:

[0043] Inputting the medical image data into a feature extraction structure to obtain a low-level feature map and a plurality of high-level feature maps;

[0044] Inputting the high-level feature map into a coding structure to obtain a coded feature map;

[0045] The encoded feature map and the low-level feature map are input into the decoding structure to obtain the segmentation result.

[0046] In this application, multiple surgical simulation modules are set up to correspond to different types of orthopedic surgeries, thereby greatly increasing the coverage of the preoperative planning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a schematic diagram of the structure of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0048] Figure 2 This is a schematic diagram of the interface of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0049] Figure 3 This is a schematic diagram of the structure of the hip joint simulation module of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0050] Figure 4 This is a schematic diagram of the structure of the knee joint simulation module of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0051] Figure 5 This is a schematic diagram of the structure of the spinal joint simulation module of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0052] Figure 6 This is a schematic diagram of the structure of the sports medicine simulation module of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0053] Figure 7 This is a schematic diagram of the structure of the trauma surgery simulation module of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0054] Figure 8This is the model architecture diagram of the bone segmentation module of the artificial intelligence preoperative planning system for all orthopedic diseases in this application;

[0055] Fig. 9 This is the model architecture diagram of the feature extraction structure of the bone segmentation module of the artificial intelligence preoperative planning system for multiple orthopedic diseases in this application. DETAILED DESCRIPTION

[0056] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below in conjunction with the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0057] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by technicians in the field to which this application belongs.

[0058] The present application embodiment provides an artificial intelligence preoperative planning system for multiple orthopedic diseases. Figure 1-Figure 9 As shown, Figure 1 , Figure 2 As shown, it is a structural diagram of the artificial intelligence preoperative planning system for all orthopedic diseases, and the artificial intelligence preoperative planning system for all orthopedic diseases includes:

[0059] Import module, used to import medical imaging data;

[0060] A three-dimensional reconstruction module is used to establish a corresponding three-dimensional image model based on medical imaging data;

[0061] The bone segmentation module is used to segment bones in medical imaging data;

[0062] And at least three of the following modules:

[0063] Hip joint simulation module, used to simulate hip joint surgery process and determine the preoperative planning plan of hip joint surgery based on medical imaging data of hip joint;

[0064] Knee joint simulation module, used to simulate the knee joint surgery process and determine the preoperative planning plan of knee joint surgery based on the medical imaging data of the knee joint;

[0065] The spinal joint simulation module is used to simulate the spinal joint surgery process and determine the preoperative planning scheme of the spinal joint surgery based on the medical imaging data of the spinal joint;

[0066] Sports medicine simulation module, used to simulate the movement process of the knee joint based on the medical imaging data of the knee joint;

[0067] The trauma surgery simulation module is used to simulate the trauma surgery process and determine the preoperative planning plan for trauma surgery based on the medical imaging data of the trauma site.

[0068] In this application, multiple surgical simulation modules are set up to correspond to different types of orthopedic surgeries, thereby greatly increasing the coverage of the preoperative planning system.

[0069] In one embodiment, in combination Figure 3 As shown, the hip joint simulation module includes a first key point recognition submodule, a first template measurement submodule, a first display submodule and a first guide plate submodule;

[0070] The first key point identification submodule is used to identify key points of the femur and acetabulum;

[0071] The first module measurement submodule is used to measure or calculate the anteversion angle, abduction angle, coverage rate of the acetabular cup prosthesis template, the difference in leg length and offset after surgery, and the amount and position of osteotomy at the proximal femur;

[0072] The first display submodule is used to display an image of the prosthesis template fitting with the bone, and to adjust it according to the operator's operation;

[0073] The first guide plate submodule is used to output the planned osteotomy amount and osteotomy position, and to derive an osteotomy guide plate model.

[0074] In one embodiment, the key points identified by the first key point identification submodule include: the center of the left femoral head, the center of the femoral diameter, the top of the lesser trochanter, the lesser trochanter, the proximal point of the femur, the distal point of the femur, the medial epicondyle and the lateral epicondyle, the center of the right femoral head, the top of the lesser trochanter, the lesser trochanter, the proximal point of the femur, and the distal point of the femur.

[0075] The process of identifying the center of the femoral head by the first key point identification submodule is as follows:

[0076] Based on the centroid formula of the plane image corresponding to the three-dimensional model, the pixel coordinates of the center point of the femoral head in the femoral area in the three-dimensional image are obtained; the pixel coordinates are converted into image coordinates; and the rotation center position of the femoral head is determined.

[0077] In this application, the identification of other key points can be performed according to actual conditions, or through a pre-trained recognition model, and the specific process will not be repeated in this application.

[0078] In the present application, the first module measurement submodule is used to calculate and determine the replaced acetabular cup prosthesis:

[0079] According to the rotation center position of the femoral head, first size information is obtained to determine second size information according to the first size, wherein the first size information is size information corresponding to the femoral head, and the second size information is size information corresponding to the acetabular cup prosthesis model.

[0080] The first module measurement submodule is also used to determine the type and placement position of the femoral stem prosthesis model:

[0081] The medullary cavity area is determined based on the identified femoral area and cortical bone area; the midpoint coordinates of each medullary cavity layer in the medullary cavity area are calculated, and based on the midpoint coordinates, all center points are fitted with a straight line to determine the anatomical axis of the medullary cavity; based on the anatomical axis of the medullary cavity and the axis of the femoral neck, the angle value of the femoral neck-shaft angle is calculated; based on the angle value, the medullary cavity area and the rotation center position of the femoral head, the type and placement position of the femoral stem prosthesis model are determined.

[0082] In the present application, after the type and placement position of the prosthesis are determined, the osteotomy position and amount are directly determined by placing it at the corresponding position and performing Boolean operations.

[0083] The first display submodule is used to display the image of the prosthesis template and the bone fitting, to display the femoral stem prosthesis and the acetabular prosthesis, and to cover the corresponding position of the hip joint through the prosthesis; if the placement position of the acetabular cup prosthesis and the placement position of the femoral stem prosthesis meet the preset position requirements, they can be directly output through the first guide submodule, or, they can be adjusted according to the operator's operation and then output.

[0084] Among them, the preset position requirement of the acetabular cup prosthesis can be, for example, that after the acetabular cup prosthesis is placed in the acetabular socket, the coverage rate of the acetabular socket is greater than 75%, and the preset position requirement of the femoral stem prosthesis can be, for example, that after the femoral stem prosthesis is placed in the medullary cavity, the angle between the long axis of the femoral stem prosthesis and the long axis of the femur is less than or equal to 3°.

[0085] In one embodiment, in combination Figure 4 As shown, the knee joint simulation module includes a second key point recognition submodule, a second template measurement submodule, a second display submodule and a second guide plate submodule;

[0086] The second key point identification submodule is used to identify key points on the femur and tibia;

[0087] The second module measurement submodule is used to measure or calculate the amount of distal femoral osteotomy, the amount of posterior femoral condyle osteotomy, and the amount of tibial plateau osteotomy, and to calculate the sum of the amount of osteotomy on the medial side of the distal femur and the medial side of the tibial plateau, the sum of the amount of osteotomy on the lateral side of the distal femur and the lateral side of the tibial plateau, the sum of the amount of osteotomy on the medial side of the posterior femoral condyle and the medial side of the tibial plateau, and the sum of the amount of osteotomy on the lateral side of the posterior femoral condyle and the lateral side of the tibial plateau;

[0088] The second display submodule is used to display an image of the prosthesis template fitting with the bone, and to adjust it according to the operator's operation;

[0089] The second guide plate submodule is used to output the planned osteotomy amount and osteotomy position, and to export the osteotomy guide plate model.

[0090] In one embodiment, the key points identified by the second key point identification submodule include: medial epicondyle, lateral epicondyle, distal medial side, distal lateral side, medial side of posterior condyle, lateral side of posterior condyle, anterior condyle, proximal point of femur and distal point of femur, medial point of tibial plateau, lateral point of plateau, medial 1 / 3 of tibial tuberosity, attachment point of posterior cruciate ligament, proximal point of tibia and distal point of tibia.

[0091] In this application, any one or more neural network models of MTCNN, locnet, Pyramid Residual Module, Densenet, hourglass, resnet, SegNet, Unet, R-CNN, Fast R-CNN, Faster R-CNN, R-FCN, and SSD are used to identify the key points.

[0092] In the present application, the second module measurement submodule is also used to calculate key axes, which include the femoral anatomical axis, the femoral mechanical axis, the tibial anatomical axis and the tibial mechanical axis, as well as the tibiofemoral angle and the distal femoral angle.

[0093] Preferably, the key axis also includes any one or more of the condylar line, the posterior condylar line, the tibial knee joint line, the femoral sagittal axis, the femoral knee joint line, and the posterior femoral condylar angle.

[0094] In the present application, the second display submodule is used to display the image of the prosthesis template and the skeleton fitting, and adjust it according to the operator's operation; wherein the displayed image of the prosthesis template and the skeleton fitting can be scaled, rotated along any axis, and moved.

[0095] Thus, the three-dimensional structure of the bones is displayed more intuitively, allowing doctors (or other medical staff) to observe images of the bone structure from multiple angles and levels.

[0096] In one embodiment, in combination Figure 5 As shown, the spinal joint simulation module includes a third key point recognition submodule, a third template measurement submodule, a third display submodule and a third guide plate submodule;

[0097] The third key point identification submodule is used to identify the key points of each vertebra of the spine;

[0098] The third module measurement submodule is used to measure or calculate the size and setting position of the screw module, the fusion device template, the titanium rod template and the cross-connection template;

[0099] The third display submodule is used to display the image of the prosthesis template fitting with the bone, and to adjust it according to the operation of the operator;

[0100] The third guide plate submodule is used to output the planned screw module, fusion cage template, titanium rod template and cross-connection template positions, and export the spinal guide plate model.

[0101] Among them, the size and setting position of the screw can be generated by a pre-trained neural network model;

[0102] For example: the preprocessed image data is input into the neural network model in the measurement submodule of the third module to obtain the screw entry point and direction, the screw diameter and screw length are calculated based on the screw entry point and screw direction, and the suitable screw model is determined in the prosthesis database based on the screw diameter and screw length.

[0103] Among them, the size and setting position of the screw can also be determined by the setting method:

[0104] Left side nail entry point: Identify the midpoint of the inferior articular process and draw a vertical line 3mm outside of the point. Identify the base of the transverse process and draw a horizontal line 1 / 3 above the base. Take the intersection of the two lines. Right side nail entry point: Identify the midpoint of the transverse process and draw a horizontal line through the point. Identify the outer edge of the superior articular process and draw a vertical line through the outer edge of the superior articular process. Take the intersection of the two lines. Nail entry direction: In the transverse plane, the left and right screws are each tilted 35° inward. In the sagittal plane, the left and right screws are parallel to the upper and lower end plates.

[0105] Screw diameter: Divide the pedicle into n planes perpendicular to the direction of screw insertion. Calculate the pedicle diameter of each plane. When the minimum pedicle diameter of all planes is ≤4.0mm, the screw diameter is 3.5mm. When the minimum pedicle diameter of all planes is greater than 4.0mm, the screw diameter is 4.0mm.

[0106] Screw length: From the entry point, draw a straight line along the direction of the screw until it breaks through the cortical bone at the front edge of the vertebral body. Calculate the distance from the entry point to the breakthrough point. 80% of this distance is the screw insertion depth. Calculate the value a = screw length - screw insertion depth * 80%. When a>0 and the value of a is the smallest, it is the optimal screw length.

[0107] Among them, the size and setting position of the titanium rod are generated: the width of the U-shaped structure at the tail of the screw is calculated, and the titanium rod curve is fitted. The optimal titanium rod width is calculated based on the width of the U-shaped structure at the tail of the screw, and the suitable titanium rod model is determined in the prosthesis database based on the optimal titanium rod width.

[0108] The size and setting position of the fusion device are generated by calculating the length and width of the intervertebral space between two adjacent vertebrae, and determining the appropriate fusion device model in the prosthesis database based on the length and width of the intervertebral space between two adjacent vertebrae.

[0109] The principle of calculating the length and width of the intervertebral space and matching the best model is:

[0110] Length: Identify the front and back edges of the vertebral body. The distance between the two points is the length of the intervertebral space.

[0111] Width: Identify the front edge and the back edge of the vertebral body, and draw a straight line connecting the two points. Draw a vertical line at 1 / 2 of the straight line. Take the two points where the vertical line intersects the vertebral body. The distance between the two points is the intervertebral width.

[0112] Matching the best model: Calculate the value a = intervertebral disc length - cage length. When a>0 and a is the smallest value, it is the best cage length. Calculate the value b = intervertebral disc width / 2 - cage width. When b>0 and b is the smallest value, it is the best cage width.

[0113] The size and setting position of the cross-connection are generated by calculating the required cross-connection length based on the titanium rod fitting curve, and determining the suitable cross-connection model in the prosthesis database based on the required cross-connection length;

[0114] Marking a certain length of the left titanium rod and the right titanium rod to obtain a first mark point and a second mark point, and connecting the first mark point and the second mark point with a straight line;

[0115] Within the length range of the titanium rod, the straight line is translated up and down, and during the translation process, the straight line always connects the left titanium rod and the right titanium rod;

[0116] The length interval of the straight line is the length range of the horizontal connection, the maximum length is taken as m, and the calculated value a=horizontal connection model (ie, the length of the horizontal connection)-m. When a≥0 and the value of a is the smallest, it is the best horizontal connection model.

[0117] In one embodiment, in combination Figure 6 As shown, the sports medicine simulation module includes a fourth key point recognition submodule, a fourth template measurement submodule and a fourth display submodule;

[0118] The fourth key point identification submodule is used to identify the key points of the femur and tibia;

[0119] The fourth module measurement submodule is used to measure or calculate the length and expansion and contraction change of the bone tunnel at different flexion angles of the knee joint;

[0120] The fourth display submodule is used to display the dynamic effects of the knee joint at different flexion angles and the effects of bone tunnel changes.

[0121] Among them, the key points identified by the fourth key point identification submodule are two condylar points (medial condylar point E, lateral condylar point F), point A on the medial and posterior superior side of the lateral femoral condyle, point B on the lateral edge of the anterior femoral condyle, point C at the anterior and medial 1 / 3 of the tibial intercondylar eminence, and point D on the medial edge of the tibial tuberosity, a total of 6 points.

[0122] The fourth module measurement submodule is also used to calculate the length expansion and contraction difference between point A on the medial posterior upper side of the lateral condyle of the femur and point C at the 1 / 3 of the anterior medial side of the intercondylar eminence of the tibia when the femur and tibia are at different angles; and determine the positions of point A on the medial posterior upper side of the lateral condyle of the femur and point C at the 1 / 3 of the anterior medial side of the intercondylar eminence of the tibia with the smallest length expansion and contraction difference as the optimized positions, thereby achieving optimization of key points.

[0123] Specifically, key point position optimization:

[0124] Calculate the shortest position of AC length expansion and contraction change when the femur and tibia are at 0°, 90°, and 120°. Points A and C are the initial points for point recognition.

[0125] Keeping the tibia still, rotate the femur 90° and 120° around the condylar line and calculate the distance between points A and C.

[0126] A. C point optimization method:

[0127] Taking the initial points A and C identified by the point recognition network as the center of the sphere, find a point set on the bone surface with a radius of 5 mm, and take one point from each of the A and C point sets to form N AC point combinations.

[0128] Calculate the difference in AC length expansion and contraction when each AC point is at 0°, 90°, and 120°.

[0129] Repeat the second step, count the difference of AC length expansion and contraction when each group of AC points are at 0°, 90°, and 120°, and find the AC point pair corresponding to the minimum expansion and contraction value of AC point.

[0130] Output the optimized coordinates of points A and C.

[0131] The fourth module measurement submodule is also used for: performing bone tunnel planning based on the knee joint CT image, and obtaining three-section bone tunnel planning results on the tibial side, ligament side, and femoral side.

[0132] Specifically, bone tunnel planning:

[0133] The knee bone tunnel is divided into three sections: tibial side, ligament, and femoral side. The femoral side has a tunnel formed by points A and B, the tibial side has a tunnel formed by points C and D, and the ligament has a tunnel formed by points A and C.

[0134] A knee bone tunnel is formed based on the tibial side, ligament, and femoral side.

[0135] The fourth display submodule is used to display: the AC length expansion and contraction changes of the femur and tibia at 0°, 90°, and 120°.

[0136] In one embodiment, in combination Figure 7 As shown,

[0137] A trauma surgery simulation module, including a fifth key point recognition submodule, a fifth template measurement submodule, a fifth display submodule and a fifth guide plate submodule;

[0138] The fifth key point identification submodule is used to identify key points of the femur and acetabulum;

[0139] The fifth module measurement submodule is used to measure or calculate the size and setting position of the main nail module, the tail cap template, the spiral blade template and the interlocking screw template;

[0140] The fifth display submodule is used to display the image of the prosthesis template fitting with the bone, and to adjust it according to the operation of the operator;

[0141] The fifth guide plate submodule is used to output the planned positions of the main nail module, tail cap template, spiral blade template and interlocking screw template.

[0142] Wherein, the trauma is intertrochanteric fracture of the femur.

[0143] The fifth module measurement submodule is also used to: repair the broken bone at the trauma position and determine the rotation matrix of the broken bone; and determine the size and setting position of the instrument template according to the repaired trauma position.

[0144] The step of repairing the broken bone at the trauma position and determining the rotation matrix of the broken bone includes:

[0145] Fitting the proximal broken bone, determining the fitted center point of the femoral head and the center point of the femoral neck, and determining the axis of the femoral head and neck; the proximal broken bone is the femoral head, and the distal broken bone is the femoral shaft;

[0146] The distal fractured bones were sampled and fitted to obtain the axis of the femoral shaft;

[0147] Extract the fracture surface point set of the proximal fractured bone and the fracture surface point set of the distal fractured bone;

[0148] The rotation matrix of the femoral head is determined based on the fracture surface point set, the femoral shaft axis, and the femoral head and neck axis.

[0149] Determining the rotation matrix of the femoral head according to the fracture surface point set and the axis line of the femoral shaft and the axis line of the femoral head and neck includes:

[0150] Generate the initial rotation matrix of the femoral head;

[0151] The neck-shaft angle was calculated based on the femoral shaft axis and the adjusted femoral head-neck axis;

[0152] The anteversion angle was calculated based on the adjusted femoral head and neck axis and the reference plane;

[0153] Iterate the rotation matrix until the neck-shaft angle and the forward tilt angle meet a preset angle range;

[0154] Calculate the point set distance according to the fracture surface point set of the femoral shaft and the fracture surface point set of the femoral head after iteration;

[0155] The rotation matrix is ​​iterated again until the distance between the point sets is minimized.

[0156] In the present application, the neck-shaft angle and the anteversion angle are first constrained within a preset angle range by adjusting the rotation matrix; then, under the premise of constraining the neck-shaft angle and the anteversion angle, the rotation matrix is ​​fine-tuned to minimize the point set distance. At this time, the fracture surface of the femoral shaft and the fracture surface of the femoral head should fit exactly, thereby achieving the position repair of the femoral fracture.

[0157] In the present application, the distance between the fracture surfaces of the femoral head and the femoral shaft is calculated to verify whether the splicing result meets the alignment splicing standard; the neck-shaft angle and the anteversion angle are calculated to verify whether the splicing result meets the alignment splicing standard; thereby achieving accurate alignment splicing and alignment splicing of the femur and realizing accurate repair.

[0158] In this application, the point set distance is the Euclidean distance between two fracture surface point sets. The ICP (Iterative Closest Point) algorithm can be used to find the point pair set with the minimum distance matching to further optimize the rotation matrix.

[0159] In this application, in the image showing the fitting of the instrument template to the bone, the process of obtaining the initial planned positions of the main nail module, the tail cap template, the spiral blade template and the locking screw template is as follows:

[0160] Main nail planning: Determine the best insertion path for the main nail between the proximal and distal femoral fractures. Usually, the main nail is inserted through the opening of the greater trochanter or the lower edge of the acetabulum. Determine the length and diameter of the main nail so that it can span the entire fracture area and provide adequate support. Align the femoral axis: Align the centerline of the main nail with the femoral axis to ensure that the main nail can pass through the medullary cavity without destroying the bone anatomy.

[0161] Tail cap planning: The tail cap is usually installed at the tail end of the main nail and should be completely aligned with the main nail. Make sure the position of the tail cap does not cause damage to soft tissues (such as tendons or nerves). Choose a tail cap with the appropriate diameter and length according to the anatomical structure of the bone to avoid the tail cap protruding from the bone or entering the joint cavity.

[0162] Planning of spiral blade: Choose a suitable insertion path in the femoral head and neck area to ensure that the spiral blade can fully enter the femoral head and cover a larger bone area. The insertion angle should be consistent with the axis of the femoral neck to avoid damaging the articular surface of the femoral head. Choose the length of the spiral blade according to the size of the femoral head and the length of the femoral neck. Usually, ensure that the blade head is at least 5 mm away from the articular surface of the femoral head to avoid cutting the articular surface.

[0163] Positioning of interlocking screws: Install one or more interlocking screws at the distal end of the main nail. The screw position should be precisely aligned with the locking hole on the main nail.

[0164] In one embodiment, in combination Figure 4 As shown, the knee joint simulation module also includes a unicompartmental submodule, which is connected to a second key point recognition submodule, a second template measurement submodule, a second display submodule and a second guide plate submodule. When the unicompartmental submodule is in an activated state, the second key point recognition submodule, the second template measurement submodule, the second display submodule and the second guide plate submodule only process operations corresponding to a single knee joint.

[0165] In the present application, the preoperative planning of the unicompartmental knee is similar to that of the knee joint, except that, in the preoperative planning of the knee joint, both knee joints are identified and planned; whereas, in the preoperative planning of the unicompartmental knee, only the knee joint on the side to be treated is identified and planned.

[0166] In the present application, a unicompartmental submodule connected to each submodule in the knee joint simulation module is set up, so that when performing unicompartmental preoperative planning, the second key point identification submodule, the second template measurement submodule, the second display submodule and the second guide plate submodule are controlled to only process the knee joint on one side.

[0167] In this application, the specific preoperative planning of unicompartmental arthroplasty can refer to the preoperative planning of the knee joint, and the details will not be repeated here.

[0168] In one embodiment, in combination Figure 8 As shown, the skeleton segmentation module is provided with a pre-trained segmentation model to segment the input medical image data, including:

[0169] Inputting the medical image data into a feature extraction structure to obtain a low-level feature map and a plurality of high-level feature maps;

[0170] Inputting multiple high-level feature maps into the encoding structure to obtain an encoded feature map;

[0171] The encoded feature map and the low-level feature map are input into the decoding structure to obtain the segmentation result.

[0172] In one embodiment, in combination Figure 8 As shown, the multiple high-level feature maps are input into the encoding structure to obtain the encoding feature map, including:

[0173] Inputting multiple high-level feature maps into multiple branches set in parallel to obtain corresponding multiple branch feature maps;

[0174] Perform convolution processing on multiple branch feature maps respectively to obtain a global feature map;

[0175] The global feature maps of some branches are spliced ​​to obtain a spliced ​​feature map;

[0176] Perform channel attention processing and convolution processing on the concatenated feature map to obtain an attention feature map;

[0177] The attention feature map is concatenated with the global feature map of the remaining branches, and spatial attention and 1×1 convolution are performed to obtain the encoded feature map.

[0178] In one embodiment, in combination Figure 8 As shown, the encoding feature map and the low-level feature map are input into the decoding structure to obtain the segmentation result, including:

[0179] Inputting multiple branches in parallel to the low-level feature map to obtain corresponding multiple decoding branch feature maps;

[0180] Perform convolution processing on multiple branch feature maps respectively to obtain a decoded global feature map;

[0181] The decoded global feature maps of some branches are concatenated and convolved to obtain a convolution feature map;

[0182] The convolution feature map and the decoded global feature map of the remaining branches are concatenated and processed by convolution layers to obtain a decoded convolution map;

[0183] The decoded convolution map and the upsampled encoded feature map are fused, convolved, and upsampled to obtain the segmentation result.

[0184] Combination Figure 8 As shown, there are five high-level feature maps. For the five feature maps, five parallel branches are set in the encoding structure, which correspond to the five extracted feature maps of the input respectively; among them, the first branch is spliced ​​after 1×1 convolution and 1×1 convolution; the second branch is spliced ​​after 3×3 convolution and 1×1 convolution; the third branch is spliced ​​after 3×3 convolution and 1×1 convolution; the fourth branch is spliced ​​after 3×3 convolution and 1×1 convolution; the fifth branch is spliced ​​after 1×1 convolution.

[0185] Among them, in the specific splicing process, the first branch, the second branch, the third branch and the fourth branch are directly spliced, and after the channel attention processing, they are spliced ​​with the fifth branch, and then the spatial attention processing is performed.

[0186] In this application, channel attention and spatial attention are extracted separately, which can make full use of the characteristics of both and capture information of different dimensions respectively, avoid the interference of information of different dimensions, make the model more focused on information extraction in a specific dimension, and improve the performance of the model.

[0187] In this application, the 1×1 convolution of the first branch, the 3×3 convolution of the second branch, the third branch, and the fourth branch are hollow convolutions, and the expansion rates are different; in this way, the extracted feature maps are strengthened by different hollow convolutions, and then spliced ​​and dynamically enhanced or weakened through the channel attention mechanism to distinguish the different features of the bones to be segmented and the surrounding tissues; the feature is extracted again and spliced ​​with the remaining branches, and then the spatial attention effect is used, that is, according to the attention of different areas, the features of the area of ​​interest are strengthened, the contrast between the bone area and the surrounding tissues is improved, and the features are easier to distinguish. In this way, the shape and texture of the bones can be highlighted, and the boundaries between the bones and the surrounding tissues can be segmented more clearly and completely.

[0188] In this application, by setting a dilated convolution, the convolution kernel has a larger receptive field to cover a larger range of input image areas, which helps to capture a wider range of contextual information and make the extracted features more semantic.

[0189] It should be noted here that the 3×3 convolutions in different branches have the same structure but different parameters.

[0190] Combination Figure 8 As shown, there is one low-level feature map, which is the feature map first processed by the feature extraction structure and is also one of the high-level feature maps.

[0191] Combination Figure 8 As shown in the figure, there are five parallel branches in the decoding structure, and the low-level feature maps are input into the five branches at the same time; among them, the first branch is spliced ​​after 1×1 dilated convolution; the second branch is spliced ​​after 3×3 dilated convolution; the third branch is spliced ​​after 3×3 dilated convolution; the fourth branch is spliced ​​after 3×3 dilated convolution; and the fifth branch is spliced ​​after 1×1 dilated convolution. Among them, the structures of 1×1 dilated convolution and 3×3 dilated convolution in different branches are the same, but the parameters and expansion rates are different.

[0192] In the specific splicing process, the first branch, the second branch, the third branch and the fourth branch are directly spliced, and after a 3×3 convolution process, they are spliced ​​with the fifth branch, and then a 1×1 convolution process is performed to obtain a decoded convolution graph.

[0193] Among them, after the encoded feature map is upsampled, it is fused with the decoded convolution map, and the fusion process can be a splicing process or other processes.

[0194] In the present application, convolution processing and upsampling processing are performed after splicing to obtain the segmentation result. After the upsampling processing, the corresponding segmentation result map can be obtained through a 1x1 convolution layer or a Softmax or Sigmoid activation function.

[0195] It should be noted that in this application, the feature maps to be spliced ​​and fused are of the same size; if the sizes are different, they can be corrected by adding additional 1×1 convolutions and other size adjustment processes. This application will not go into details.

[0196] In one embodiment, in combination Fig. 9 As shown, the medical image data is input into a feature extraction structure to obtain a low-level feature map and multiple high-level feature maps, including:

[0197] Perform 7×7 convolution on medical image data to obtain a large convolution feature map;

[0198] Perform global extraction processing on the large convolution feature map to obtain a global extraction feature map;

[0199] After concatenating the global extraction feature map and the large convolution feature map, a first high-level feature map is obtained; the first high-level feature map is also a low-level feature map;

[0200] Performing pooling processing on the first high-level feature map to obtain a pooled feature map;

[0201] Perform a combination of 1×1 convolution, 3×3 convolution, and 1×1 convolution on the pooled feature map to obtain a combined feature map;

[0202] Concatenate the pooled feature map and the combined feature map to obtain a second high-level feature map;

[0203] Performing a combination of 1×1 convolution, 3×3 convolution, and 1×1 convolution on the second high-level feature map to obtain a high-level combined feature map;

[0204] Concatenate the high-level combined feature map and the combined feature map to obtain a third high-level feature map;

[0205] Perform a combination of 1×1 convolution, 3×3 convolution, and 1×1 convolution on the third high-level feature map to obtain a high-level three-combination feature map;

[0206] The high-level combined feature map and the high-level three-combination feature map are concatenated to obtain a fourth high-level feature map;

[0207] The fourth high-level feature map is pooled and fully connected to obtain a fifth high-level feature map.

[0208] In this application, by directly performing 7×7 convolution processing on the medical image data, richer contextual information and more large objects or large-scale features in the medical image data are directly obtained, so that more global information can be captured at an early level, and the generated low-level feature map contains more global information, thereby improving the global characteristics.

[0209] In the present application, the first high-level feature map is simultaneously input into the decoding structure as a low-level feature map, and the first high-level feature map is reused to improve the global characteristics of the low-level feature map while maintaining the multi-view and multi-size of the encoding structure.

[0210] In this application, by combining convolution and residual splicing, on the one hand, key features are continuously extracted, and on the other hand, through continuous residuals, the global characteristics and quasi-low-level features in subsequent multiple high-level feature maps are maintained, thereby greatly improving the corresponding characteristics in the output high-level feature maps.

[0211] In one implementation, performing global extraction processing on the large convolution feature map to obtain a global extracted feature map includes:

[0212] Input the large convolution feature map into a 1×1 convolution layer set in parallel to obtain a first extraction map, a second extraction map, and a third extraction map respectively; wherein the first extraction map and the third extraction map are in HW×N format, and the second extraction map is in N×HW format;

[0213] Multiply the first extracted image and the second extracted image to obtain a multiplied feature image;

[0214] Perform softmax-like processing on the multiplied feature map to obtain the multiplication coefficient;

[0215] Multiplying the multiplication coefficient with the third extraction map to obtain a coefficient multiplication map;

[0216] Add the coefficient multiplication map and the large convolution feature map to obtain the global extraction feature map.

[0217] In the present application, the large convolution feature map is divided into three branches for 1×1 convolution processing. Through the convolution, the dimensions of the first branch and the third branch are adjusted to the HW×N format, and the dimension of the second branch is adjusted to the N×HW format; then the first branch is multiplied by the third branch to obtain a feature map with a dimension of HW×HW, and a coefficient is obtained through a softmax-like processing; the coefficient is multiplied with the third branch, and the multiplied feature map is added to the large convolution feature map to obtain a global extraction feature map.

[0218] In this application, the dimensions of the first branch and the second branch are reversed through 1×1 convolution, thereby realizing multiplication between feature maps; through this multiplication, long-range dependencies are captured; by setting coefficients, the dependencies are embedded in the feature maps; and on this basis, the original input is added and a residual connection is used to embed the global extraction into the model without destroying the parameters.

[0219] In this application, long-range dependencies between features are captured through global extraction, so that the model can extract global features, making up for the defect that the current model mainly extracts local features.

[0220] In one implementation, the activation function of the SoftMax-like operation is:

[0221]

[0222] Among them, SoftMax1(x) is the activation function, x i 、x j is the element in the input vector, i and j are the element numbers.

[0223] In this application, SoftMax is a mathematical function that is usually used to convert a set of arbitrary real numbers into real numbers representing probability distributions. It is essentially a normalization function that can convert a set of arbitrary real values ​​into probability values ​​between [0,1], because SoftMax converts them to values ​​between 0 and 1, so they can be interpreted as probabilities. If one of the inputs is small or negative, SoftMax turns it into a small probability, and if the input is large, it turns it into a large probability, but it will always remain between 0 and 1.

[0224] However, for the standard SoftMax function, since the input is mapped between 0 and 1, and the sum of all output values ​​is 1, this means that even if some input values ​​are very small, they will have a non-zero output value after being processed by the SoftMax function. This will also cause the noise to be amplified, resulting in the final output result being affected by greater noise.

[0225] In this application, a 1 is added to the denominator of the SoftMax-like function; this change means that when the input value is very small, their output value can be closer to zero. This allows the corresponding output to tend to zero when there is no valuable information to add, thus greatly reducing unnecessary noise.

[0226] It should be noted that the segmentation model in this application (the conventional model is only used to segment part of the bones and the structure is too simple) is used to segment the entire orthopedic AI, not only for the segmentation of the knee joint, but also for the segmentation of the hip joint, spine and trauma part. In this application, the accurate segmentation of bones in complex human bodies is achieved by retaining global features multiple times and setting low-level feature maps and high-level feature maps.

[0227] The specific training process of the segmentation model will not be repeated in this application.

[0228] The terms mentioned in this application are all conventional orthopedic terms, and the explanations of each term are as follows:

[0229] Femoral anatomical axis: centerline of the femoral shaft.

[0230] Femoral mechanical axis: one end is located at the center of the hip joint, and the other end is located at the center of the femur's knee joint (the apex of the femoral intercondylar notch).

[0231] Tibia anatomical axis: centerline of the tibial shaft.

[0232] Tibial mechanical axis: one end point is located at the center of the tibial knee joint (the center of the intercondylar spine), and the other end point is located at the center of the tibial ankle joint (the midpoint of the line connecting the lateral cortical bones of the inner and outer ankles).

[0233] Condylar line: The line connecting the highest point of the medial femoral condyle and the lateral femoral condyle.

[0234] Posterior condylar line: The line connecting the lowest points of the medial and lateral posterior condyles of the femur.

[0235] Femoral knee joint line: the line connecting the lowest points of the distal femur.

[0236] Tibial knee joint line: The line connecting the medial low point and lateral high point of the tibial plateau.

[0237] Femoral sagittal axis: the line connecting the center of the posterior cruciate ligament insertion and the medial edge of the tibial tuberosity.

[0238] Tibiofemoral angle (mTFA): The angle between the mechanical axis of the femur and the mechanical axis of the tibia.

[0239] Distal femoral angle: the angle between the mechanical axis of the femur and the anatomical axis of the femur.

[0240] The posterior condylar angle (PCA) is the angle between the projection lines of the femoral condylar line and the posterior condylar line in the cross section.

[0241] In addition, the terms "first", "second", and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", and "third" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0242] In this application, unless otherwise clearly specified and limited, the terms "disposed", "connected", "connected", "connected" and the like should be understood in a broad sense, for example, it can be an electrical connection / communication connection, or an electrical coupling / communication coupling, or integrated; it can be directly connected, or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0243] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradicting each other.

[0244] It should be noted that in the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some embodiments, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this description.

[0245] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An artificial intelligence preoperative planning system for multiple orthopedic diseases, characterized by: include: Import module, used to import medical imaging data; A three-dimensional reconstruction module is used to establish a corresponding three-dimensional image model based on medical imaging data; The bone segmentation module is used to segment bones in medical imaging data; And at least three of the following modules: Hip joint simulation module, used to simulate hip joint surgery process and determine the preoperative planning plan of hip joint surgery based on medical imaging data of hip joint; Knee joint simulation module, used to simulate the knee joint surgery process and determine the preoperative planning plan of knee joint surgery based on the medical imaging data of the knee joint; The spinal joint simulation module is used to simulate the spinal joint surgery process and determine the preoperative planning scheme of the spinal joint surgery based on the medical imaging data of the spinal joint; Sports medicine simulation module, used to simulate the movement process of the knee joint based on the medical imaging data of the knee joint; The trauma surgery simulation module is used to simulate the trauma surgery process and determine the preoperative planning plan for trauma surgery based on the medical imaging data of the trauma site.

2. The artificial intelligence preoperative planning system for multiple diseases in orthopedics according to claim 1 is characterized in that: The hip joint simulation module includes a first key point recognition submodule, a first template measurement submodule, a first display submodule and a first guide plate submodule; The first key point identification submodule is used to identify key points of the femur and acetabulum; The first module measurement submodule is used to measure or calculate the anteversion angle, abduction angle, coverage rate of the acetabular cup prosthesis template, the difference in leg length and offset after surgery, and the amount and position of osteotomy at the proximal femur; The first display submodule is used to display an image of the prosthesis template fitting with the bone, and to adjust it according to the operator's operation; The first guide plate submodule is used to output the planned osteotomy amount and osteotomy position, and to derive an osteotomy guide plate model.

3. The artificial intelligence preoperative planning system for all orthopedic diseases according to claim 2 is characterized in that: The key points identified by the first key point identification submodule include: the center of the left femoral head, the center of the femoral diameter, the top of the lesser trochanter, the lesser trochanter, the proximal point of the femur, the distal point of the femur, the medial epicondyle and the lateral epicondyle, the center of the right femoral head, the top of the lesser trochanter, the lesser trochanter, the proximal point of the femur, and the distal point of the femur.

4. The artificial intelligence preoperative planning system for all orthopedic diseases according to any one of claims 1 to 3, characterized in that: The knee joint simulation module includes a second key point recognition submodule, a second template measurement submodule, a second display submodule and a second guide plate submodule; The second key point identification submodule is used to identify key points on the femur and tibia; The second module measurement submodule is used to measure or calculate the amount of distal femoral osteotomy, the amount of posterior femoral condyle osteotomy, and the amount of tibial plateau osteotomy, and to calculate the sum of the amount of osteotomy on the medial side of the distal femur and the medial side of the tibial plateau, the sum of the amount of osteotomy on the lateral side of the distal femur and the lateral side of the tibial plateau, the sum of the amount of osteotomy on the medial side of the posterior femoral condyle and the medial side of the tibial plateau, and the sum of the amount of osteotomy on the lateral side of the posterior femoral condyle and the lateral side of the tibial plateau; The second display submodule is used to display an image of the prosthesis template fitting with the bone, and to adjust it according to the operator's operation; The second guide plate submodule is used to output the planned osteotomy amount and osteotomy position, and to export the osteotomy guide plate model.

5. The artificial intelligence preoperative planning system for all orthopedic diseases according to claim 4 is characterized in that: The key points identified by the second key point identification submodule include: medial epicondyle, lateral epicondyle, distal medial side, distal lateral side, medial side of posterior condyle, lateral side of posterior condyle, anterior condyle, proximal point of femur and distal point of femur, medial point of tibial plateau, lateral point of plateau, medial 1 / 3 of tibial tuberosity, attachment point of posterior cruciate ligament, proximal point of tibia and distal point of tibia.

6. The artificial intelligence preoperative planning system for all orthopedics diseases according to any one of claims 1 to 3, characterized in that: A spinal joint simulation module, including a third key point recognition submodule, a third template measurement submodule, a third display submodule and a third guide plate submodule; The third key point identification submodule is used to identify the key points of each vertebra of the spine; The third module measurement submodule is used to measure or calculate the size and setting position of the screw module, the fusion device template, the titanium rod template and the cross-connection template; The third display submodule is used to display the image of the prosthesis template fitting with the bone, and to adjust it according to the operation of the operator; The third guide plate submodule is used to output the planned screw module, fusion cage template, titanium rod template and cross-connection template positions, and export the spinal guide plate model.

7. The artificial intelligence preoperative planning system for all orthopedics diseases according to any one of claims 1 to 3, characterized in that: A sports medicine simulation module, including a fourth key point recognition submodule, a fourth template measurement submodule and a fourth display submodule; The fourth key point identification submodule is used to identify the key points of the femur and tibia; The fourth module measurement submodule is used to measure or calculate the length and expansion and contraction change of the bone tunnel at different flexion angles of the knee joint; The fourth display submodule is used to display the dynamic effects of the knee joint at different flexion angles and the effects of bone tunnel changes.

8. The artificial intelligence preoperative planning system for all orthopedics diseases according to any one of claims 1 to 3, characterized in that: A trauma surgery simulation module, including a fifth key point recognition submodule, a fifth template measurement submodule, a fifth display submodule and a fifth guide plate submodule; The fifth key point identification submodule is used to identify key points of the femur and acetabulum; The fifth module measurement submodule is used to measure or calculate the size and setting position of the main nail module, the tail cap template, the spiral blade template and the interlocking screw template; The fifth display submodule is used to display the image of the prosthesis template fitting with the bone, and to adjust it according to the operation of the operator; The fifth guide plate submodule is used to output the planned positions of the main nail module, tail cap template, spiral blade template and interlocking screw template.

9. The artificial intelligence preoperative planning system for all orthopedic diseases according to claim 4 is characterized in that: The knee joint simulation module also includes a unicompartmental submodule, which is connected to a second key point recognition submodule, a second template measurement submodule, a second display submodule and a second guide plate submodule. When the unicompartmental submodule is in an activated state, the second key point recognition submodule, the second template measurement submodule, the second display submodule and the second guide plate submodule only process operations corresponding to a single knee joint.

10. The artificial intelligence preoperative planning system for all orthopedic diseases according to any one of claims 1-3, characterized in that: The bone segmentation module is provided with a pre-trained segmentation model, and the segmentation of the input medical image data includes: Inputting the medical image data into a feature extraction structure to obtain a low-level feature map and a plurality of high-level feature maps; Inputting the high-level feature map into a coding structure to obtain a coded feature map; The encoded feature map and the low-level feature map are input into the decoding structure to obtain the segmentation result.

Images