Neural network-based computer-aided orthopedic optimization method for six-axis external fixators

Abstract

The present application relates to a kind of six-axis external fixator computer-aided orthopedic optimization method based on neural network, the method includes: based on coordinate parameter and length parameter, inverse kinematics and forward kinematics are solved to six-axis external fixator, the length parameter and relative pose of six screw rods are obtained.Reference bone segment is determined, with the broken end of reference bone segment as reference point, and pose transformation and relationship matrix are constructed according to deformity parameter and installation parameter, wherein, deformity parameter and installation parameter are measured by clinical diagnosis.Safety orthopedic speed is determined, and the change sequence of parameter in deformity process is calculated, the pose transformation matrix corresponding to each group of deformity parameter is obtained, and the spatial position relationship between distal ring and proximal ring is deduced.By optimizing kinematics calculation method, the length of telescopic screw on six-axis external fixator is calculated, and adjustment scheme is completed.The method is optimized by kinematics algorithm, and the pose change trajectory of external fixator and bone segment is optimized to be closer to ideal trajectory.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bone deformity correction, and in particular to a six-axis external fixator computer-aided orthopedic optimization method based on a neural network. BACKGROUND

[0002] For the clinical treatment of fractures and deformity diseases of the human foot and ankle, due to the complex joint function and tissue environment of the foot and ankle, an external fixation treatment method is often used. At present, in order to achieve the synchronous adjustment of multi-plane foot and ankle deformity and complete the correction of translation and torsion deformity, various types of six-axis fixators have been widely used and adopted.

[0003] When using this type of external fixator for treatment, by inputting clinical diagnosis and treatment parameters to the matching computer-aided software, a fine and accurate orthopedic prescription scheme can be obtained. However, when adjusting the six screws of the six-axis fixator, the instrument configuration will form several intermediate states, so the treatment trajectory is locally nonlinear, and the linear and uniform stable change of the fixed ring pose cannot be truly realized. In order to minimize the patient's pain as much as possible, it is necessary to realize the minimum cumulative deviation of the nonlinear treatment trajectory on the basis of meeting the orthopedic treatment target, that is, to make the actual treatment trajectory closer to the ideal trajectory. In actual operation, it is embodied as the optimization of the screw adjustment sequence.

[0004] At present, in the clinical aspect, the computer-aided adjustment scheme design method used for foot and ankle orthopedics does not consider the above optimization of the trajectory, and the patient will suffer pain caused by several times of external fixator stretching. At the same time, the research on different foot and ankle orthopedic treatment modes and the kinematics principle of the mechanism of this type of six-axis external fixator is insufficient, and the research involving the optimization process does not consider the attitude change of the bone segment in the treatment trajectory, and the calculation is relatively complex, which increases the time cost and is not conducive to further application. SUMMARY

[0005] Therefore, it is necessary to provide a six-axis external fixator computer-aided orthopedic optimization method based on a neural network, which meets the correction effect of different types of deformities and shortens the treatment time on the premise of alleviating the discomfort of patients.

[0006] A six-axis external fixator computer-aided orthopedic optimization method based on a neural network, the six-axis external fixator is composed of a proximal ring, a distal ring and six screws for connecting the proximal ring and the distal ring, wherein the proximal ring is used to fix the proximal bone segment of a broken bone, and the distal ring is used to fix the distal bone segment of the broken bone.

[0007] The six-axis external fixator computer-aided orthopedic optimization method comprises the following steps:

[0008] Based on the coordinate parameters and the length parameters, inverse kinematics and forward kinematics of the six-axis external fixator are solved to obtain length parameters and relative poses of the six screws;

[0009] A reference bone segment is determined, a broken end of the reference bone segment is taken as a reference point, and a pose transformation and a relationship matrix are constructed according to deformity parameters and installation parameters, wherein the deformity parameters and the installation parameters are measured by clinical diagnosis;

[0010] A safe orthopedic speed is determined, and a change sequence of the parameters in the deformity process is calculated to obtain a pose transformation matrix corresponding to each group of deformity parameters, and a spatial position relationship between the distal ring and the proximal ring is derived;

[0011] By using the optimization kinematics calculation method, the length of the telescopic screw on the six-axis external fixator is calculated to complete the adjustment scheme.

[0012] In one embodiment, based on the coordinate parameters and the length parameters, the inverse kinematics and the forward kinematics of the six-axis external fixator are solved, including:

[0013] The geometric centers of the proximal ring and the distal ring are taken as origins to design a proximal ring coordinate system S p and a distal ring coordinate system S d , the proximal ring coordinate system is assumed to be a reference coordinate system, and coordinate parameters P(x d ,y d ,z d ,α d ,β d ,γ d ) are obtained;

[0014] Based on the proximal ring coordinate system S p , a first vector relationship is established in the distal ring coordinate system S d to realize rotation transformation of a spatial vector pose, and an expression of the first vector relationship is:

[0015] Based on the first vector relationship, a second vector relationship of the screw connection hinge point is derived, the first vector in the distal ring coordinate system S d is converted into a second vector in the proximal ring coordinate system S p , and an expression of the second vector relationship is:

[0016] A third vector on the proximal ring connection hinge point in the S p is subtracted from the second vector to obtain a first length, and an expression of the first length is:

[0017] wherein, denotes S d the first vector connecting the hinge point of the middle distal ring to the telescopic screw; denotes S p the second vector connecting the hinge point of the middle distal ring to the telescopic screw; denotes S p the third vector connecting the hinge point of the middle proximal ring; T is a rotation transformation matrix; R and t are rotation and movement components in the matrix, respectively; and i is the screw number.

[0018] In one of the embodiments, the inverse kinematics and the forward kinematics of the six-axis external fixator based on the coordinate parameters and the length parameters further include:

[0019] A first multi-variable nonlinear equation is established based on the relationship between the first initial length of the screw and the second vector and the third vector on the proximal ring, and the expression of the first multi-variable nonlinear equation is:

[0020] A cumulative deviation amount objective function between the sub-node pose change trajectory and the ideal pose change trajectory is established based on the deviation relationship between the actual pose change trajectory and the theoretical trajectory of the distal ring, and the objective function is taken as the evaluation basis, and the expression of the objective function is:

[0021]

[0022] wherein, ΔT represents the cumulative deviation amount of the position change trajectory, ΔR represents the cumulative deviation amount of the attitude change trajectory, l i (T) represents the distance of each sub-node in the sub-node position change trajectory to the theoretical position change trajectory, l j (R) represents the distance of each sub-node in the sub-node attitude change trajectory to the theoretical attitude change trajectory, l0(T) and l0(R) represent the straight line lengths of the theoretical position and attitude trajectories between two nodes, respectively.

[0023] In one of the embodiments, the inverse kinematics and the forward kinematics of the six-axis external fixator based on the coordinate parameters and the length parameters further include:

[0024] A multi-layer perceptron deep neural network is established in combination with the objective function to solve the first multi-variable nonlinear equation to obtain the intermediate node length of the screw, and the multi-layer perceptron neural network is composed of multiple layers of neurons and includes an input layer, an output layer and a hidden layer, and the expression is:

[0025]

[0026] In the formula, n is the input quantity of the layer, m is the number of neurons contained in the layer, the input and output layers are both 6, the hidden layers are both 20, ω ij and b j are the interlayer connection weight and threshold value respectively, x and y are the input and output of the layer respectively, f is an activation function, the activation functions of the hidden layers are all set to ReLu functions, and the activation function of the output layer is set to an identity function;

[0027] The multi-layer perception neural network input is the initial length, the output is the initial pose calculation result, and the loss expression is:

[0028]

[0029] In the formula, T i =(x i -x i ') 2 +(y i -y i ') 2 +(z i -z i ') 2 , R i =(α i -α i ') 2 +(β i -β i ') 2 +(γ i -γ i ') 2 , s represents the number of single samples when the network is trained by using small batches of samples, x i , y i , z i , α i , β i and γ i represent ideal pose parameters in sample data, x i ', y i ', z i ', α i ', β i ' and γ i ' represent calculated pose parameters output by the network;

[0030] The absolute value difference comparison expression is:

[0031] |ΔL i |=|L in -L cal |<δ(i=1~6)。

[0032] In one of the embodiments, the inverse kinematics solving and the forward kinematics solving of the six-axis external fixator based on the coordinate parameters and the length parameters further comprise:

[0033] Based on the initial pose calculation result, inverse kinematics solving is performed again to obtain an intermediate node length of the screw rod;

[0034] Based on the intermediate node length, an error length is obtained by performing absolute value difference comparison with the initial length, if the error length is less than a first threshold value, the initial pose calculation result is output as an ideal pose calculation result, if the error length is greater than the first threshold value, error compensation is performed, and the expression of the error compensation is:

[0035] L in =L in +ΔL i ;

[0036] In the formula, ΔL i (i=1~6) is the error length, L in is the initial length, L cal is the intermediate node length, and δ is the first threshold value.

[0037] In one of the embodiments, the deformity parameters are represented as D(D x ,D y ,D z ,D rx ,D ry ,D rz ), and the installation parameters are represented as M(M x ,M y ,M z ,M rx ,M ry ,M rz ), which are used to describe the difference between the foot and ankle deformity cases and the installation of the proximal ring of the six-axis external fixator on the patient's limb or foot.

[0038] In one of the embodiments, the reference bone segment is determined, the broken end of the reference bone segment is taken as a reference point, and the pose transformation and the relationship matrix are constructed according to the deformity parameters and the installation parameters, comprising:

[0039] According to the deformity parameters and the installation parameters, pose transformation matrices T D and T M are constructed, which respectively describe the spatial pose relationship of the second bone segment and the proximal ring relative to the first bone segment;

[0040] Based on the six-axis external fixator forward kinematics, the matrix T H is obtained by combining the screw rod length, which is used to represent the spatial pose relationship of the distal ring relative to the proximal ring.

[0041] According to the pose transformation matrix T D , T M and T H , the matrix T R can be derived to describe the spatial position relationship of the distal ring relative to the second bone segment, and the relationship expression is:

[0042] T R = T D -1 × T M × T H .

[0043] In one embodiment, the safe orthopedic speed is determined, and the change sequence of the parameter in the deformity process is calculated to obtain the pose transformation matrix corresponding to each group of deformity parameters, and the spatial position relationship between the distal ring and the proximal ring is derived, including:

[0044] The calculation expression of the change sequence is:

[0045]

[0046]

[0047] In the formula, D i represents the deformity state after i changes, D P represents the treatment target, Numbers represents the number of changes of the deformity parameters required to complete the treatment target, j is used to select different indicators in a group of deformity parameters, and CorrectSpeed represents the safe orthopedic speed;

[0048] The derivation expression of the spatial position relationship between the distal ring and the proximal ring is:

[0049]

[0050] In one embodiment, the length of the telescopic screw rod on the six-axis external fixator is calculated by the optimization kinematics calculation method to complete the adjustment scheme, including:

[0051] Based on the inverse kinematics of the six-axis external fixator, the lengths of the six screw rods connected to the proximal ring are calculated and the orthopedic adjustment scheme is obtained.

[0052] The above-mentioned neural network-based six-axis external fixator computer-aided orthopedic optimization method, the six-axis external fixator is composed of a proximal ring, a distal ring, and six screw rods for connecting the proximal ring and the distal ring, wherein the proximal ring is used to fix the proximal bone segment of the fractured bone, and the distal ring is used to fix the distal bone segment of the fractured bone. The six-axis external fixator computer-aided orthopedic optimization method comprises the following steps: based on the coordinate parameters and the length parameters, inverse kinematics solving and forward kinematics solving are performed on the six-axis external fixator to obtain the length parameters and the relative pose of the six screw rods. A reference bone segment is determined, the broken end of the reference bone segment is taken as a reference point, and a pose transformation and a relationship matrix are constructed according to the deformity parameters and the installation parameters measured by clinical diagnosis. A safe orthopedic speed is determined, and the change sequence of the parameters in the deformation process is calculated to obtain the pose transformation matrix corresponding to each group of deformity parameters, and the spatial position relationship between the distal ring and the proximal ring is derived. Finally, the length of the telescopic screw rod on the six-axis external fixator is calculated by an optimized kinematics calculation method, and the adjustment scheme is completed. This method optimizes the kinematics algorithm, calculates the best order of each group of screw rod adjustment when each group of screw rod is adjusted, and obtains the bone segment treatment trajectory with the minimum cumulative offset of the theoretical trajectory, so as to improve the treatment experience of the patient and better meet the correction effect of different types of deformity diseases, so that the pose change trajectory of the external fixator and the bone segment fixed thereto in the treatment process is optimized to be closer to the ideal trajectory, the treatment time is shortened under the premise of maintaining the treatment experience of the patient. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 The neural network-based six-axis external fixator computer-aided orthopedic optimization method flowchart of one embodiment in the present application;

[0054] Figure 2 The six-axis external fixator basic structure schematic diagram of one embodiment in the present application;

[0055] Figure 3 The position change trajectory schematic diagram of the mediation process of one embodiment in the present application;

[0056] Figure 4 The neural network structure schematic diagram of one embodiment in the present application;

[0057] Figure 5 The screw rod length error compensation flowchart of one embodiment in the present application;

[0058] Figure 6 The clinical diagnosis and treatment measurement parameter diagram of one embodiment in the present application;

[0059] Figure 7 The conventional ankle configuration mode schematic diagram of one embodiment in the present application;

[0060] Figure 8A schematic diagram of a butt joint series connection configuration mode of an embodiment in the present application;

[0061] Figure 9 A schematic diagram of a butt joint series connection configuration mode of an embodiment in the present application;

[0062] Figure 10 A schematic diagram of a parallel connection configuration mode of an embodiment in the present application.

[0063] In the figure: 100, screw; 110, proximal ring; 111, proximal bone segment; 120, distal ring; 121, distal bone segment; 200, full ring; 210, U-shaped ring; 300, tibia; 310, foot; 320, forefoot; 330, hind foot; 400, two-thirds ring; 500, additional ear plate. DETAILED DESCRIPTION

[0064] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0065] As shown in Figures 1 to 6 , in one embodiment, a neural network-based computer-aided orthopedic optimization method for a six-axis external fixator, the six-axis external fixator is composed of a proximal ring 110, a distal ring 120, and six screws 100 for connecting the proximal ring 110 and the distal ring 120, wherein the proximal ring 110 is used to fix the proximal bone segment 111 of the broken bone, and the distal ring 120 is used to fix the distal bone segment 121 of the broken bone. The method includes the following steps:

[0066] Step S110, based on the coordinate parameters and the length parameters, inverse kinematics and forward kinematics of the six-axis external fixator are solved to obtain the length parameters and relative poses of the six screws.

[0067] Specifically, the coordinate system definition includes designing a proximal ring coordinate system S p and a distal ring coordinate system S d with the geometric centers of the two rings as the origins, respectively, while assuming that the proximal ring coordinate system is the reference coordinate system. The relative coordinate parameters P(x d ,y d ,z d ,α d ,β d ,γ dBased on X-ray anteroposterior and lateral views or clinical examination measurements, describe the relative position and orientation of the control ring coordinate system relative to the reference ring coordinate system along the six spatial degrees of freedom.

[0068] Based on the proximal ring coordinate system S p In the far-end ring coordinate system S d A first vector relationship is established to realize the rotation transformation of the spatial vector pose. The expression of the first vector relationship is:

[0069] Based on the first vector relationship, the second vector relationship of the screw connection hinge point is derived, and the far-end ring coordinate system S is... d The first vector in the vector is transformed into the proximal annular coordinate system S. p The expression for the relationship between the second vector and the second vector is:

[0070] Combined with S p The third vector at the proximal-to-mid-end ring connection hinge point is subtracted from the second vector, and the absolute value is taken to obtain the first length. The expression for the first length is:

[0071] In the formula, S represents d The first vector of the hinge point connecting the telescopic screw on the mid-to-far end ring; S represents p The second vector of the hinge point connecting the telescopic screw on the mid-to-far end ring; S represents p The third vector connecting the hinge point on the proximal end ring; T is the rotation transformation matrix; R and t are the rotation and translation components in the matrix, respectively; i is the screw number.

[0072] It should be noted that, In the formula, c represents cos; s represents sin.

[0073] Based on the initial lengths of the six screws and the relationship between the second and third vectors on the near-end ring, a multivariate nonlinear equation is established. The expression for the multivariate nonlinear equation is as follows:

[0074]

[0075] Based on the deviation relationship between the actual pose change trajectory and the theoretical trajectory of the far-end ring, an objective function for the cumulative deviation between the pose change trajectory of the sub-node and the ideal pose change trajectory is established, and the objective function is used as the evaluation criterion. The expression of the objective function is as follows:

[0076]

[0077] In the formula, ΔT represents the cumulative deviation amount of the position change trajectory, ΔR represents the cumulative deviation amount of the attitude change trajectory, l i (T) represents the distance of each sub-node in the sub-node position change trajectory to the theoretical position change trajectory, l j (R) represents the distance of each sub-node in the sub-node attitude change trajectory to the theoretical attitude change trajectory, l0(T) and l0(R) represent the straight line lengths of the theoretical position and attitude trajectories between two nodes, respectively.

[0078] In combination with the objective function, a multilayer perceptron deep neural network is established to solve a multivariate nonlinear equation to obtain the intermediate node length of the screw rod. The multilayer perceptron neural network is composed of multiple neurons, including an input layer, an output layer and a hidden layer, and its expression is as follows:

[0079]

[0080] In the formula, n is the input quantity of the layer, m is the number of neurons contained in the layer, the input and output layers are both 6, the hidden layers are both 20, ω ij and b j are the interlayer connection weights and thresholds, respectively, x and y are the input and output of the layer, respectively, and f is an activation function. The activation function of the hidden layer is set to a ReLu function, and the activation function of the output layer is set to an identity function.

[0081] The input of the multilayer perceptron neural network is the initial length, the output is the initial pose calculation result, and the loss expression is as follows:

[0082]

[0083] In the formula, T i =(x i -x i ') 2 +(y i -y i ') 2 +(z i -z i ') 2 , R i =(α i -α i ') 2 +(β i -β i ') 2 +(γ i -γ i ') 2 , s represents the number of single samples when the network is trained by using small batches of samples, x i , y i , z i , α ix i y i z i α i β i γ i x i y i z i

[0084] The absolute value difference comparison expression is:

[0085] |ΔL in |=|L cal -L in |<δ(i=1~6)。

[0086] Based on the initial pose calculation result, inverse kinematics is solved again to obtain the intermediate node length of the screw rod. Based on the intermediate node length, the absolute value difference comparison is made with the initial length to obtain the error length. If the error length is less than the first threshold value, the initial pose calculation result is output as the ideal pose calculation result. If the error length is greater than the first threshold value, error compensation is performed. The expression of error compensation is:

[0087] L in =L i +ΔL i 。

[0088] In the formula, ΔL in (i=1~6) is the error length, L cal is the initial length, L x is the intermediate node length, and δ is the first threshold value.

[0089] In step S120, a reference bone segment is determined, the broken end of the reference bone segment is taken as a reference point, and a pose transformation and a relationship matrix are constructed according to a deformity parameter and an installation parameter, wherein the deformity parameter and the installation parameter are measured by clinical diagnosis.

[0090] Specifically, the deformity parameter is represented as D(D y ,D z ,D rx ,D ry ,D rz ), and the installation parameter is represented as M(M x ,M y ,M z ,M rx ,M ry ,M rz ), which are used to describe the difference between the foot and ankle deformity cases and the installation of the proximal ring of the six-axis external fixator on the patient's limb or foot and ankle. ​

[0091] According to the deformity parameters and the installation parameters, a pose transformation matrix T is constructed D and T M respectively describe the spatial pose relationship of the proximal bone segment and the proximal ring relative to the distal bone segment.

[0092] Based on the forward kinematics of the six-axis external fixator, the matrix T is obtained by combining the screw rod length H , which is used to represent the spatial pose relationship of the distal ring relative to the proximal ring. According to the pose transformation matrix T D , T M and T H , the matrix T can be obtained R to describe the spatial position relationship of the distal ring relative to the second bone segment, and the relationship expression is:

[0093] T R = T D -1 × T M × T H .

[0094] Step S130, determine the safe orthopedic speed, and calculate the change sequence of the deformity parameters to obtain the pose transformation matrix corresponding to each group of deformity parameters, and deduce the spatial position relationship between the distal ring and the proximal ring.

[0095] Specifically, the calculation expression of the change sequence is:

[0096]

[0097]

[0098] In the formula, D i represents the deformity state after i changes, D P represents the treatment target, Numbers represents the number of changes of the deformity parameters required to complete the treatment target, j is used to select different indicators in a group of deformity parameters, and CorrectSpeed represents the safe orthopedic speed.

[0099] The spatial position relationship between the distal ring and the proximal ring is deduced as follows:

[0100]

[0101] Step S140, calculate the length of the telescopic screw rod on the six-axis external fixator by optimizing the kinematics calculation method, and complete the adjustment scheme.

[0102] Specifically, based on the inverse kinematics of the six-axis external fixator, the lengths of the six screw rods connected to the proximal ring are calculated and the orthopedic adjustment scheme is obtained.

[0103] The above-mentioned neural network-based six-axis external fixator computer-aided orthopedic optimization method, the six-axis external fixator is composed of a proximal ring, a distal ring and six screw rods for connecting the proximal ring and the distal ring, wherein the proximal ring is used to fix the proximal bone segment of the fractured bone, and the distal ring is used to fix the distal bone segment of the fractured bone. The six-axis external fixator computer-aided orthopedic optimization method comprises the following steps: based on the coordinate parameters and the length parameters, the inverse kinematics and the forward kinematics of the six-axis external fixator are solved to obtain the length parameters and the relative pose of the six screw rods. The reference bone segment is determined, the broken end of the reference bone segment is taken as the reference point, and the pose transformation and the relationship matrix are constructed according to the deformity parameters and the installation parameters measured by clinical diagnosis. The safe orthopedic speed is determined, and the change sequence of the parameters in the deformity process is calculated to obtain the pose transformation matrix corresponding to each group of deformity parameters, and the spatial position relationship between the distal ring and the proximal ring is derived. Finally, the length of the telescopic screw rod on the six-axis external fixator is calculated by the optimization kinematics calculation method, and the adjustment scheme is completed. This method optimizes the kinematics algorithm, calculates the best order of each group of screw rod adjustment when each group of single-step adjustment of the six screw rods is performed, and obtains the bone segment treatment trajectory with the minimum cumulative offset of the corresponding theoretical trajectory, so as to improve the treatment experience of the patient and better meet the correction effect of different types of deformity diseases, so that the pose change trajectory of the external fixator and the bone segment fixed thereto in the treatment process is optimized to be closer to the ideal trajectory, the treatment time is shortened on the premise of maintaining the treatment experience of the patient.

[0104] As shown in Figures 7 to 10 , a neural network-based six-axis external fixator computer-aided adjustment optimization method is provided, which includes four different types of application schemes as specific embodiments, and the four types of external fixation supports include a conventional ankle configuration mode, a bevel joint series configuration mode, a butt joint series configuration mode and a parallel configuration mode, which are all deformations of the basic structure in Figure 2 .

[0105] Those skilled in the art should recognize that the following four different types of embodiments are only used to illustrate the main content of the algorithm of the present application, and are not used as a limitation of the present application. As long as the changes and other modifications of the above embodiments are within the scope of the spirit and essence of the algorithm of the present application, they will fall within the scope of the technical disclosure of the present application.

[0106] Embodiment 1:

[0107] The conventional ankle configuration mode can treat diseases such as clubfoot deformity by adjusting the dorsiflexion angle. In the conventional ankle configuration mode, the six-axis external fixator generally comprises a full ring 200, a U-shaped ring 210, and a telescopic screw group composed of six screw rods 100, wherein the full ring 200 is fixedly connected to the tibia 300 of the patient, the U-shaped ring 210 is fixedly connected to the foot 310 and the opening faces forward, and the telescopic screw group is sequentially connected to the two rings in the order shown in the figure.

[0108] In embodiment 1, the tibia 300 and the foot 310 can be reference bone segments, and the deformity parameters D measured in clinical diagnosis and treatment and the installation parameters M can respectively describe the relative pose relationship between the foot 310 and the tibia 300 of the patient and between the reference bone segments and the fixed reference ring. Combined with the lengths of the six telescopic screw rods and according to the method described in the main content of the present application, the corresponding orthosis adjustment scheme can be calculated and obtained.

[0109] Embodiment 2:

[0110] The oblique joint series connection configuration is mainly formed by splicing two groups of six-axis external fixators, comprising a full ring 200, two two-thirds rings 400, and two groups of telescopic screw groups composed of six screw rods 100, wherein the two two-thirds rings 400 are fixedly connected to the tibia 300 and the forefoot 320, respectively, and the full ring 200 is fixedly connected to the hindfoot 330, and are sequentially connected to the full ring 200 as shown in the figure.

[0111] In embodiment 2, the correction of the hindfoot 330 and the forefoot 320 relative to the tibia 300 can be simultaneously achieved, so that two sets of orthosis adjustment schemes can be obtained by two independent calculations to guide the doctor or the patient to slowly adjust the two groups of six-axis external fixators, respectively, to simultaneously correct the ankle joint deformity and midfoot deformity including clubfoot deformity. For the proximal group of six-axis external fixation orthosis, the tibia 300 and the midfoot and hindfoot can be used as reference bone segments during the adjustment scheme calculation process, and for the distal six-axis external fixation orthosis, the forefoot and the midfoot and hindfoot can be used as reference bone segments, and according to the method described in the main content of the present application, the corresponding orthosis adjustment scheme can be calculated and obtained.

[0112] Embodiment 3:

[0113] The standard butt series configuration is composed of two groups of six-axis external fixators, including a full ring 200, a two-thirds ring 400, two groups of telescopic screw rods composed of six screw rods 100, and two U-shaped rings 210. The full ring 200 is fixedly connected to the tibia 300, the two-thirds ring 400 is fixedly connected to the forefoot, the U-shaped ring 210 is fixed on the hind foot 330, is parallel to the axis direction of the tibia 300, and is connected to the U-shaped ring 210, the U-shaped ring 210 is fixed on the tibia 300, and the screw rod 100 is divided into two groups, one of which is connected to the full ring 200, and the other is connected to the two-thirds ring 400, and is connected to the two U-shaped rings 210 in the order shown in the figure.

[0114] In Example 3, deformities between the forefoot 320 and the hind foot 330 and ankle deformities can be corrected at the same time, such as correcting deformed diabetic feet. As in the oblique series configuration mode, the corresponding adjustment scheme of each group of six-axis external fixators can be obtained by independent calculation, and in the calculation process, the distal end of the tibia 300 and the forefoot 320 and the hind foot 330 can be used as reference bone segments, and the corresponding orthopedic adjustment scheme is calculated according to the method described in the main content of the application.

[0115] Example 4:

[0116] The parallel configuration mode includes a full ring 200, two two-thirds rings 400, two groups of telescopic screw rods composed of six screw rods 100, and an additional ear plate 500. The full ring 200 is fixedly connected to the tibia 300, the two two-thirds rings 400 are respectively fixedly connected to the forefoot 320 and the hind foot 330, the additional ear plate 500 is connected to the full ring, the screw rod 100 is divided into two groups, one of which is connected to the full ring 200, and the other is connected to one of the two-thirds rings 400, the additional ear plate 500 is connected to the full ring 200, and is connected to the other two-thirds ring 400 and the additional ear plate 500 in the order shown in the figure.

[0117] In Example 4, the patient's forefoot 320 and hind foot 330 are taken as two independent units and are orthopedically treated by using two groups of six-axis external fixators. One group connects the full ring 200 to the two-thirds ring 400 through six screw rods 100 for treating the hind foot 330 deformity, and in the calculation of the adjustment scheme, the tibia 300 and the hind foot 330 can be used as reference bone segments. The other group connects the additional ear plate 500 to the two-thirds ring 400 through the other six screw rods 100 for treating the forefoot 320 deformity, and in the calculation of the adjustment scheme of the six-axis external fixator, the forefoot 320 is usually selected as the reference bone segment, and the corresponding orthopedic adjustment scheme is calculated according to the method described in the main content of the application.

[0118] Any combination of the technical features in the above embodiments can be made, and for the sake of brevity, not all possible combinations are described above. However, any combination of features described above is deemed to be within the scope of the present disclosure, as long as such a combination is not obviously contradictory.

[0119] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the patent scope of the present application. It should be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these are within the protection scope of the present application. Therefore, the protection scope of the present application patent should be subject to the appended claims.

Claims

1. A six-axis external fixator computer-aided orthopedic optimization method, characterized by, The six-axis external fixator is composed of a proximal ring, a distal ring and six screw rods for connecting the proximal ring and the distal ring, wherein the proximal ring is used for fixing the proximal bone segment of a fractured bone, and the distal ring is used for fixing the distal bone segment of the fractured bone; The computer-aided orthopedic optimization method of the six-axis external fixator comprises the following steps: Based on the coordinate parameters and the length parameters, inverse kinematics and forward kinematics of the six-axis external fixator are solved to obtain the length parameters and relative poses of the six screw rods; A reference bone segment is determined, a reference point is taken as the broken end of the reference bone segment, and a pose transformation and a relationship matrix are constructed according to deformity parameters and installation parameters, wherein the deformity parameters and the installation parameters are measured by clinical diagnosis; A safe orthopedic speed is determined, and a change sequence of parameters in the deformation process is calculated to obtain a pose transformation matrix corresponding to each group of deformity parameters, and a spatial position relationship between the distal ring and the proximal ring is derived; By optimizing the kinematics calculation method, the length of the telescopic screw rod on the six-axis external fixator is calculated, and the adjustment scheme is completed; based on the inverse kinematics of the six-axis external fixator, the length of the six screw rods connected to the proximal ring is calculated And get the orthopedic adjustment scheme.

2. The hexaxial external fixator computer-aided orthopedic optimization method of claim 1, wherein, The inverse kinematics and forward kinematics of the six-axis external fixator based on the coordinate parameters and the length parameters comprise: With the geometric centers of the proximal and distal loops as the origin, design the proximal loop coordinate system S. p and the far-end ring coordinate system S d Assuming the proximal ring coordinate system is the reference coordinate system, obtain the coordinate parameters P(x). d ,y d ,z d ,α d ,β d ,γ d ); Based on the proximal ring coordinate system S p , a first vector relationship is established in the distal ring coordinate system S d , realizing the rotation transformation of the spatial vector pose, and the expression of the first vector relationship is: based on the first vector relationship, deducing a second vector relationship of the screw connection hinge point, converting the first vector in the distal ring coordinate system S d into a second vector in the proximal ring coordinate system S p , and an expression of the second vector relationship is: In conjunction with the S p A third vector on the medial proximal ring connection hinge point, subtract the absolute value of the second vector, get the first length, the expression of the first length is: wherein represents S d the first vector of the distal ring connecting the hinge point of the telescopic screw, represents S p the second vector of the distal ring connecting the hinge point of the telescopic screw, represents S p the third vector of the proximal ring connecting the hinge point, T is the rotation transformation matrix, R and t are the rotation and translation components in the matrix, respectively, and i is the screw number.

3. The six-axis external fixator computer-aided orthopedic optimization method of claim 2, wherein, The inverse kinematics and forward kinematics of the six-axis external fixator based on the coordinate parameters and the length parameters further comprise: Based on the relationship between the first initial length of the screw rod and the second vector and the third vector on the proximal ring, a first multi-element nonlinear equation is established, and the expression of the first multi-element nonlinear equation is: Based on the deviation relationship between the actual pose change trajectory and the theoretical trajectory of the distal ring, a cumulative deviation amount objective function between the sub-node pose change trajectory and the ideal pose change trajectory is established, and the objective function is taken as the evaluation basis, and the expression of the objective function is: where ΔT represents the cumulative deviation amount of the position change trajectory, ΔR represents the cumulative deviation amount of the attitude change trajectory, l i (T) represents the distance of each of the sub-node position change trajectories to the theoretical position change trajectory, l j (R) represents the distance of each of the sub-node attitude change trajectories to the theoretical attitude change trajectory, l0(T) and l0(R) represent the straight line lengths of the theoretical position and attitude trajectories between the two nodes, respectively.

4. The six-axis external fixator computer-aided orthopedic optimization method of claim 3, wherein, The inverse kinematics and forward kinematics of the six-axis external fixator based on the coordinate parameters and the length parameters further comprise: Combined with the objective function, a multi-layer perceptron deep neural network is established to solve the first multi-element nonlinear equation to obtain the intermediate node length of the screw rod, and the multi-layer perceptron neural network is composed of multiple layers of neurons, including an input layer, an output layer and a hidden layer, and the expression is: In the formula, n is the input quantity of the layer, m is the number of neurons contained in the layer, the input and output layers are both 6, the hidden layers are both 20, ω ij and b j are respectively the interlayer connection weight and threshold value, x and y are respectively the input and output of the layer, and f is an activation function, the activation functions of the hidden layers are all set to ReLu functions, and the activation function of the output layer is set to an identity function. In the formula, n is the input quantity of the layer, m is the number of neurons contained in the layer, the input and output layers are both 6, the hidden layers are both 20, ω ij and b j are respectively the interlayer connection weight and threshold value, x and y are respectively the input and output of the layer, and f is an activation function, the activation functions of the hidden The input of the multi-layer perceptron neural network is the initial length, the output is the initial pose calculation result, and the loss expression is: In the formula, T i = (x i -x i ′) 2 + (y i -y i ′) 2 + (z i -z i ′) 2 , R i = (α i -α i ′) 2 + (β i -β i ′) 2 + (γ i -γ i ′) 2 , s represents the number of single samples when the network is trained by using small batch samples, x i , y i , z i , α i , β i , and γ i in the sample data represent ideal pose parameters, x i ', y i ', z i ', α i ', β i ', and γ i ' represent calculated pose parameters output by the network; The absolute value difference comparison expression is: | ΔL i | = |L in - L cal | < δ (i = 1 ~ 6); The inverse kinematics and forward kinematics of the six-axis external fixator based on the coordinate parameters and the length parameters further comprise: Based on the initial pose calculation result, inverse kinematics is solved again to obtain the intermediate node length of the screw rod; Based on the intermediate node length, an absolute value difference comparison is made with the initial length to obtain an error length, if the error length is less than a first threshold value, the initial pose calculation result is taken as an ideal pose calculation result for output, and if the error length is greater than the first threshold value, error compensation is performed, and the expression of the error compensation is: L in = L in + ΔL i ; where ΔL i (i = 1 ~ 6) is the error length, L in is the initial length, L cal is the intermediate node length, and δ is the first threshold value.

5. The hexaxial external fixator computer assisted orthopedic optimization method of Claim 1 wherein, The deformity parameters are represented as D(D x ,D y ,D z ,D rx ,D ry ,D rz ), and the installation parameters are represented as M(M x ,M y ,M z ,M rx ,M ry ,M rz ), for describing the difference between the ankle deformity cases and the installation of the proximal ring of the six-axis external fixator on the patient's limb or ankle.

6. The six-axis external fixator computer-aided orthopedic optimization method of Claim 5, wherein, The reference bone segment is determined, a reference point is taken as the broken end of the reference bone segment, and a pose transformation and a relationship matrix are constructed according to deformity parameters and installation parameters, wherein the deformity parameters and the installation parameters are measured by clinical diagnosis; From the deformity parameters and the mounting parameters, a pose transformation matrix T is constructed D and T M , respectively, describe the spatial pose relationship of the second bone segment and the proximal ring relative to the first bone segment; Based on the forward kinematics of the six-axis external fixator, a matrix T is obtained in combination with the screw length H for representing the spatial pose relationship of the distal ring relative to the proximal ring; According to the pose transformation matrix T D , T M and T H The matrix T R can be obtained to describe the spatial position relationship of the distal ring relative to the second bone segment, and the relationship expression is: T R = T D -1 x T M x T H .

7. The six-axis external fixator computer-aided orthopedic optimization method of claim 6, wherein, The safe orthopedic speed is determined, and a change sequence of parameters in the deformation process is calculated to obtain a pose transformation matrix corresponding to each group of deformity parameters, and a spatial position relationship between the distal ring and the proximal ring is derived; The change sequence calculation expression is: where D i represents the deformed state after i changes, D P represents the treatment goal, Numbers represents the number of changes to the deformed parameters needed to complete the treatment goal, j is used to select different indicators in a set of deformed parameters, and CorrectSpeed represents the safe orthopedic speed. The spatial positional relationship derivation expression between the distal ring and the proximal ring is:

Images