A method for detecting internal deformation of vertebrae using a tomography device
Through the internal deformation detection device and improved algorithm based on medical CT equipment, the problems of long scanning time, low accuracy and low computing efficiency in the prior art are solved, and efficient and accurate internal deformation detection of vertebrae are achieved, providing safer correction plan.
Patent Information
- Application Number
- CN202210080665.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-24
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-01-24
AI Technical Summary
The existing internal deformation detection methods of vertebrae have problems such as long scanning time, high equipment cost, imbalance in computing efficiency and accuracy, and incorrect matching. It is difficult to accurately measure internal vertebrae information, which affects the effectiveness and safety of correction surgery.
Using the internal deformation detection device of vertebrae based on medical electronic computed tomography equipment and an improved algorithm, uniform external force is applied through the vertebrae loading device, and images are recorded in combination with medical CT equipment, and the minimum square is normalized using Gaussian filter and weighted correction factor zero mean The distance correlation function is image-processed, combined with the three-dimensional reverse Gaussian-Newton iterative algorithm to calculate the displacement and strain field, and visualize it to display the deformation cloud diagram of the internal deformation of the vertebrae.
It significantly shortens the scanning time, improves the computational efficiency and accuracy, reduces the number of iterations, enhances the accuracy of feature sub-block matching, provides a more scientific vertebra correction treatment plan, and reduces the risk of surgical complications.
Smart Images

Figure CN115018755B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for detecting internal deformation of vertebrae of a tomography device, and belongs to the fields of biomechanics, photomechanics and experimental mechanics. Background Art
[0002] The three-dimensional deformity of the vertebrae can seriously affect the patient's physical health. It has now developed into a common clinical disease, and the optimization of its treatment plan has always been one of the hot issues of concern in the medical and engineering community. Vertebral correction is essentially a process in which the vertebrae are forced to deform by external forces and return to their normal shape. It is a typical mechanical problem. However, in actual treatment, factors such as the patient's vertebral tolerance, age group, and various causes of disease need to be considered. Excessive correction may cause the failure of internal fixations, increase the incidence of surgical complications, and even cause nerve damage and paralysis in severe cases; while too light correction will not have a significant effect and aggravate the condition. If you want to solve these problems, the key is to accurately measure the internal deformation of the vertebrae, so as to grasp the main force points of the vertebrae and provide the best balance point for corrective surgery.
[0003] Vertebrae have the characteristics of viscoelasticity, anisotropy, and heterogeneity. They contain a large number of trabecular tissues and gaps inside, and the deformation analysis of the parts connected to them is extremely complex. This makes it difficult for traditional measurement methods to accurately measure the internal information of vertebrae. Digital volume image correlation methods combined with computed tomography (CT) equipment can measure the internal deformation of objects and provide a calculation method for internal vertebrae detection. However, there are still the following problems with the digital volume image correlation method for detecting internal vertebrae deformation: (1) Previous experiments mostly used industrial CT equipment. Although the scanned vertebrae images have a high spatial resolution, the scanning time is too long. During this period, tissue autolysis and stress relaxation occur inside the vertebrae, which have an irreversible impact on its internal mechanical properties. Secondly, industrial CT equipment is expensive and difficult for general units to afford. (2) Traditional digital volume image correlation methods are difficult to balance between computational efficiency and accuracy. Traditional fitting algorithms have high computational efficiency but low computational accuracy. The three-dimensional inverse combined Gauss-Newton method has high accuracy, but requires a certain number of iterations during the calculation process, resulting in relatively low computational efficiency. Secondly, when calculating large internal deformations of objects, the zero-mean normalized least square distance correlation function used to evaluate the similarity of sub-blocks sometimes cannot accurately distinguish the corresponding feature sub-blocks, resulting in matching errors.
[0004] Medical CT scanning is fast and very common, which can effectively suppress the adverse effects of long scanning time on changes in vertebral mechanical properties; weighted correction processing of the zero-mean normalized least square distance correlation function can improve the correct matching rate of characteristic sub-blocks, and secondly, providing more accurate initial values for iterations to the three-dimensional inverse combined Gauss-Newton method can reduce the number of iterations, thereby improving efficiency while ensuring calculation accuracy; In summary, a vertebral internal deformation detection device based on medical electronic computer tomography equipment and an improved internal deformation calculation method are needed to provide an important scientific basis for vertebral correction treatment plans. Summary of the invention
[0005] The purpose of the present invention is to provide a device for detecting internal deformation of vertebrae based on medical electronic computer tomography equipment and an improved related algorithm, which can measure the internal deformation of vertebrae under different loading states in combination with medical CT equipment, thereby obtaining the displacement field and strain field of the internal deformation of vertebrae, and visualizing the displacement field and strain field data to display the internal deformation cloud map of vertebrae; the present invention specifically comprises the following steps:
[0006] (1) A vertebral loading device is used to apply a uniform external force to the vertebrae, and images of the vertebrae before and after loading are recorded using a medical CT scan, and the images are reconstructed using a CT workstation.
[0007] (2) The vertebral image is filtered using a Gaussian filter to reduce the noise in the image. The actual area of the vertebra in the CT image is marked using image morphology, and the area is cut and used as the effective calculation area.
[0008] (3) Select a point (x 0 ,y 0 ,z 0 ) as the center, and search for the target sub-block with the most similar grayscale in the deformed image. The zero-mean normalized least square distance correlation function of the weighted correction factor is used to evaluate the grayscale similarity between the two. The coarse search algorithm is used to search for the overall pixel displacement. When the zero-mean normalized least square distance correlation function of the weighted correction factor is at the minimum value, that is, C Min ≤C, then search the results and compare the position changes between the two center points to determine the overall pixel displacement (u 0 ,v 0 ,w 0 ).
[0009] (4) Select the target sub-block with the overall point (x 0 +u 0 ,y 0 +v 0 ,z 0 +w0 ) as the center and the 12 discrete points in the 3×3×3 sub-volume block are sequentially substituted into the logarithmic ternary quadratic polynomial fitting function, and the least squares method is used to calculate the primary subvoxel displacement (u 1 ,v 1 ,w 1 ) and the primary displacement is
[0010] (5) The primary displacement value obtained in step (4) As the initial value of the iterative displacement Substitute it into the zero-mean normalized least square distance correlation function, and use the three-dimensional reverse Gauss-Newton iterative algorithm to calculate the displacement. The calculated result (u, v, w) is the final displacement value.
[0011] (6) The final displacement (u, v, w) is fitted by the local point-by-point least squares fitting method to obtain the strain field information inside the vertebra, complete the internal deformation calculation of the vertebra, and visualize the displacement field and strain field data to obtain the internal deformation field cloud map of the vertebra.
[0012] The vertebral deformation cloud map display described in the present invention can three-dimensionally visualize the data of the internal deformation calculation of the vertebra, which is convenient for intuitively displaying the relevant information of the internal deformation field of the vertebra; the vertebral deformation cloud map can display the displacement and strain in different directions, including but not limited to u-direction displacement, v-direction displacement, w-direction displacement, total displacement, normal strain, shear strain, principal strain, equivalent strain, volume strain, etc.
[0013] The vertebral loading device of the present invention includes a frame 1, a turbine loading handle 2, a vortex rod 3, a loading rod 4, a sensor 5, a connecting rod 6, a connecting head 7, a support pad 8, and a static strain gauge 9; the turbine loading handle 2 is connected to the loading rod 4 through the vortex rod 3, the sensor 5 is fixed to the lower end of the loading rod 4, the sensor 5 is connected to the static strain gauge 9, the connecting head 7 is connected to the sensor 5 through the connecting rod 6, and the support pad 8 is placed at the lower part of the frame 1; the upper part of the frame 1 is made of steel, and the left part, the right part and the lower part of the frame are made of PC plastic plate.
[0014] When using medical CT scanning, the ability of X-rays to penetrate metal is weak, which can easily cause unnecessary metal artifacts and affect the measurement accuracy, while PC plastic will not cause obvious artifacts. Therefore, the upper part of the frame is made of steel, which is convenient for connecting devices such as the turbine loading handle, and does not need to be placed in the medical CT scanning chamber. The left, right and lower parts of the frame are connected by 3 cm thick PC plastic plates to facilitate clear imaging of the vertebrae; the turbine loading handle combined with the loading rod can provide uniform pressure, which is transmitted to the vertebrae through the connecting rod and the connecting head; the sensor is an S-shaped double-hole sensor, which has the advantages of high sensitivity and strong anti-eccentricity ability, and is suitable for measuring smaller load values; the loading connector and the support pad are made of POM plastic. The POM material has good rigidity and hardness, and will not deform when the external force is small.
[0015] Compared with the prior art, the present invention has the following advantages and breakthrough effects:
[0016] (1) The vertebral loading device of the present invention is small in size and easy to transport. It can be directly placed in a medical CT scanning chamber for loading, thereby solving the problem that only industrial CT can be used to collect vertebral scanning images in the past, and greatly saving the time consumed in image collection.
[0017] (2) The zero-mean normalized least square distance correlation function of the weighted correction factor adopted in the present invention can better evaluate the feature points of different sub-blocks and improve the accuracy of sub-block matching.
[0018] (3) Compared with the traditional algorithm, the optimized subvoxel displacement measurement algorithm adopted in the present invention reduces the number of iterations by about 35.58% and improves the calculation efficiency by 31.88% while ensuring the same measurement accuracy (the accuracy of the displacement is 0.0060 voxel and the precise value is 0.0005 voxel), thereby achieving more efficient calculation. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 is a flow chart of the method for detecting internal vertebral deformation of the present invention;
[0020] Figure 2 is a schematic structural diagram of the vertebral loading device of the present invention;
[0021] Figure 3 It is a schematic diagram of an S-type double-hole sensor;
[0022] Figure 4 It is a schematic diagram of a resistance bridge circuit;
[0023] Figure 5 A flowchart of the method for calculating the internal deformation of vertebrae improved by the present invention;
[0024] Figure 6The cloud diagrams show the total displacement (a), volume strain (b), and equivalent strain (c) inside the vertebra measured by the present invention.
[0025] Figure 2 Middle: 1-frame; 2-turbine loading handle; 3-turbine rod; 4-loading rod; 5-sensor; 6-connecting rod; 7-connecting head; 8-support pad; 9-static strain gauge. DETAILED DESCRIPTION
[0026] In order to better describe the specific implementation scheme of the present invention, the following will be described in conjunction with actual vertebral deformation measurements and drawings, but this should not limit the scope of protection of the present invention.
[0027] The vertebral loading device of the embodiment of the present invention comprises a frame 1, a turbine loading handle 2, a vortex rod 3, a loading rod 4, a sensor 5, a connecting rod 6, a connecting head 7, a support pad 8, and a static strain gauge 9; the turbine loading handle 2 is connected to the loading rod 4 through the vortex rod 3, the sensor 5 is fixed to the lower end of the loading rod 4, the sensor 5 is connected to the static strain gauge 9, the connecting head 7 is connected to the sensor 5 through the connecting rod 6, and the support pad 8 is placed at the lower part of the frame 1; the upper part of the frame 1 is made of steel, and the left, right and lower parts of the frame are made of PC plastic plates, such as Figures 2 to 4 shown.
[0028] Example 1
[0029] A method for detecting internal deformation of vertebrae based on medical computer tomography equipment, such as Figure 1 As shown, the specific steps include:
[0030] (1) Using a vertebral loading device to apply a uniform external force to the vertebrae, the vertebral loading device is as follows: Figure 2 , Figure 3 , Figure 4 As shown; medical CT scanning is used to record the images of the vertebrae before and after loading, and the images are reconstructed using a CT workstation.
[0031] (2) The vertebral image is filtered using a Gaussian filter to reduce the noise in the image; the actual area of the vertebra in the CT image is marked using image morphology, the area is cut, and the area is used as the effective calculation area.
[0032] (3) Select a point (x 0 ,y 0 ,z 0 ) is used as the center of the reference sub-block, and the target sub-block with the most similar grayscale in the deformed image is searched for. The zero-mean normalized least square distance correlation function of the weighted correction factor is used to evaluate the grayscale similarity between the two. Figure 5 As shown in step 1:
[0033] Zero-mean normalized least squares distance correlation function with weighted correction factor:
[0034]
[0035]
[0036] Where: (x i ,y i ,z i ) and (x i +u 0 ,y i +v 0 ,z i +w 0 ) represent the coordinates of the center point of the i-th sub-block in the reference image and the deformed image, respectively, and f and g represent their corresponding grayscale values, respectively. m , g m is the average gray value of the sub-block before and after deformation; u 0 、v 0 、w 0 is the overall displacement change of the center point of the sub-block in the x, y, and z directions; w(x i ,y i ,z i ) is the weighted correction factor of the ith sub-block, where (M,M,M) is the coordinate of the central voxel point of the reference volume image.
[0037] The coarse search algorithm is used to search for the overall prime displacement. When the zero-mean normalized least square distance correlation function of the weighted correction factor is at its minimum, that is, C Min ≤C, then search the results and compare the position changes between the two center points to determine the overall pixel displacement (u 0 ,v 0 ,w 0 ).
[0038] (4) Select the target sub-block with the overall point (x 0 +u 0 ,y 0 +v 0 ,z 0 +w 0 ) as the center and the 12 discrete points in the 3×3×3 sub-volume block are sequentially substituted into the logarithmic ternary quadratic polynomial fitting function, and the coefficients corresponding to the polynomial are calculated using the least squares method to obtain the primary subvoxel displacement (u 1 ,v 1 ,w 1 ) and the primary displacement is like Figure 5As shown in step 2:
[0039] Take the logarithm of the three-variable quadratic polynomial fitting function:
[0040] Q = exp[a 0 +a 1 x+a 2 y+a 3 z+a 4 xy+a 5 xz+a 6 yz+a 7 x 2 +a 8 y 2 +a 9 z 2 ] (2)
[0041] In the above formula: Q = [InQ 1 InQ 2 … InQ 12 ] T is the zero-mean normalized minimum square distance correlation function value of the weighted correction factor obtained by the overall prime displacement search, A = [a 0 a 1 a 2 a 3 a 4 a 5 a 6 a 7 a 8 a 9 ] T is the unknown coefficient matrix, X = [1 xyz xy xz yz x 2 y 2 z 2 ] T is the coordinate of the fitting point; the 12 fitting points are sequentially brought into the logarithmic ternary quadratic fitting function, which can be expressed in a matrix as follows:
[0042]
[0043] The coefficient matrix A is solved using the least squares method:
[0044]
[0045] The extreme points of the fitting function should meet the following conditions:
[0046]
[0047] The coordinates of the extreme points of the fitting function are:
[0048]
[0049] The primary subvoxel displacement is calculated as:
[0050]
[0051] The primary displacement is:
[0052]
[0053] (5) The primary displacement value obtained in step (4) As the initial value of the iterative displacement Substitute it into the zero-mean normalized least square distance correlation function, and use the three-dimensional reverse Gauss-Newton iterative algorithm to calculate the displacement. The calculated result (u, v, w) is the final displacement value.
[0054] The zero-mean normalized least square distance correlation function is rewritten as:
[0055]
[0056] In the above formula: X=(x 0 ,y 0 ,z 0 ) T represents the coordinates of the center point of the reference sub-block, ξ=(Δx, Δy, Δz) T is the local coordinate corresponding to the center point, is the initial value of iteration, ΔP=[Δu,Δu x ,Δu y ,Δu w ,Δv,Δv x ,Δv y ,Δv w ,Δw,Δw x ,Δw y ,Δw w ] T is the iterative increment; f(X+W(ξ;ΔP)) and g(X+W(ξ;P 0 )) represent the grayscale values of the center points of the reference and deformed sub-blocks, respectively, and f m , g m is the average gray value; W(ξ; P 0 ) is expressed as a warping function, which is mainly used to describe the change in the position or shape of the sub-block, while W(ξ;ΔP) is the incremental warping function acting on the reference sub-block. The expressions of the two warping functions are as follows:
[0057]
[0058]
[0059] The first-order iterative increment Δp is:
[0060]
[0061] Among them, Δf is the gray gradient value of each voxel in the volume image; H is the Hessian matrix, which is calculated using the following formula:
[0062]
[0063] After the iteration increment ΔP is determined, the incremental warping function W(ξ; ΔP) of the reference sub-block can be determined. Then, the W(ξ; ΔP) in the reference sub-block is compared with the warping function W(ξ; P 0 ) is combined in reverse order, and the new warping function obtained can be used to update the W(ξ; P 0 ), first updated As shown below:
[0064]
[0065] The convergence conditions are:
[0066]
[0067] When the calculation result meets the preset convergence extreme value condition ||Δp||≤0.001 or the maximum number of iterations is 10, the deformation vector P = [u,u x ,u y ,u w ,v,v x ,v y ,v w ,w,w x ,w y ,w w ] T , and obtain the final displacement (u, v, w). The corresponding total displacement can be obtained by processing the displacement field. The total displacement is:
[0068]
[0069] (6) The displacement field is fitted using the point-by-point least squares fitting method to calculate the deformation field information inside the vertebra;
[0070] A calculation area with the point to be calculated as the center and (2M+1)×(2M+1)×(2M+1) individual points around it is selected as the strain calculation window; when the window is small, the displacement can be regarded as a linear distribution state in the calculation area, and the following relationship exists at this time:
[0071]
[0072] a i, b i 、c i are the first-order polynomial coefficients respectively, that is, the first-order displacement gradient component is also the strain component, which can be obtained by the least squares method.
[0073] Determination coefficient a i , b i 、c i After that, each strain component can be calculated using the following formula:
[0074]
[0075] The volume strain, principal strain and equivalent strain of the vertebra can be calculated through the strain components.
[0076] Principal strains:
[0077]
[0078] Volumetric strain:
[0079] ε v =ε x +ε y +ε z (20)
[0080] Equivalent strain:
[0081]
[0082] (7) Visualize the displacement field and strain field data to obtain the internal deformation field cloud map of the vertebra. The total displacement, volume strain, and equivalent strain of the vertebra after loading are as follows: Figure 6 As shown in Figures 6a, 6b, and 6c, it can be seen that the total displacement inside the vertebra is between 6voxel and 13voxel and is evenly distributed; the volume strain of the vertebra shows that the volume of the upper part of the vertebra increases, while the volume of the lower part decreases; the equivalent strain equates the six strain components to the strain generated under a unidirectional pressure, which is convenient for intuitively displaying the strain.
Claims
1. A method for detecting internal deformation of vertebrae using a tomography device, It is characterized in that The specific steps include: (1) applying a uniform external force to the vertebrae using a vertebral loading device, recording images of the vertebrae before and after loading using a medical CT scan, and reconstructing the images using a CT workstation; (2) Using a Gaussian filter to filter the vertebral image to reduce the noise in the image, using image morphology to mark the actual area of the vertebra in the CT image, cutting the area, and using the area as the effective calculation area; (3) Select a point (x 0 ,y 0 ,z 0 ) as the center, and search for the target sub-block with the most similar grayscale in the deformed image. The zero-mean normalized least square distance correlation function of the weighted correction factor is used to evaluate the grayscale similarity between the two. The coarse search algorithm is used to search for the overall pixel displacement. When the zero-mean normalized least square distance correlation function of the weighted correction factor is at the minimum value, that is, C Min ≤C, then search the results and compare the position changes between the two center points to determine the overall pixel displacement (u 0 ,v 0 ,w 0 ); (4) Select the target sub-block with the overall point (x 0 +u 0 ,y 0 +v 0 ,z 0 +w 0 ) as the center and the 12 discrete points in the 3×3×3 sub-volume block are sequentially substituted into the logarithmic ternary quadratic polynomial fitting function, and the least squares method is used to calculate the primary subvoxel displacement (u 1 ,v 1 ,w 1 ) and the primary displacement is (5) The primary displacement value obtained in step (4) As the initial value of the iterative displacement Substitute it into the zero-mean normalized least square distance correlation function, and use the three-dimensional reverse Gauss-Newton iterative algorithm to calculate the displacement. The calculated result (u, v, w) is the final displacement value. (6) The final displacement (u, v, w) is fitted by the local point-by-point least squares fitting method to obtain the strain field information inside the vertebra, complete the internal deformation calculation of the vertebra, and visualize the displacement field and strain field data to obtain the internal deformation field cloud map of the vertebra.
2. The method for detecting internal deformation of vertebrae by a tomography device according to claim 1, Features: The specific process of step (3) is as follows: Zero-mean normalized least squares distance correlation function with weighted correction factor: Where: (x i ,y i ,z i ) and (x i +u 0 ,y i +v 0 ,z i +w 0 ) represent the coordinates of the center point of the i-th sub-block in the reference image and the deformed image, respectively, and f and g represent their corresponding grayscale values, respectively. m , g m is the average gray value of the sub-block before and after deformation; u 0 、v 0 、w 0 is the overall displacement change of the center point of the sub-block in the x, y, and z directions; w(x i ,y i ,z i ) is the weighted correction factor of the ith sub-block, where (M,M,M) is the coordinate of the central voxel point of the reference volume image.
3. The method for detecting internal deformation of vertebrae by a tomography device according to claim 1, Features: The specific process of step (4) is as follows: Take the logarithm of the three-variable quadratic polynomial fitting function: Q=exp[a 0 +a 1 x+a 2 y+a 3 z+a 4 xy+a 5 xz+a 6 yz+a 7 x 2 +a 8 y 2 +a 9 z 2 ] Select 12 discrete points in the 3×3×3 sub-block as fitting points, Q = [InQ 1 InQ 2 … InQ 12 ] T is the zero-mean normalized minimum square distance correlation function value of the weighted correction factor obtained by the overall prime displacement search, A = [a 0 a 1 a 2 a 3 a 4 a 5 a 6 a 7 a 8 a 9 ] T is the unknown coefficient matrix, X = [1 xyz xy xz yz x 2 y 2 z 2 ] T is the coordinate of the fitting point; the 12 fitting points are sequentially brought into the logarithmic ternary quadratic fitting function, which can be expressed in a matrix as follows: The coefficient matrix A is solved using the least squares method: The extreme points of the fitting function should meet the following conditions: The coordinates of the extreme points of the fitting function are: The primary subvoxel displacement is calculated as: The primary displacement is: 。 4. The method for detecting internal deformation of vertebrae by a tomography device according to claim 1, Features: The specific process of step (5) is as follows: The zero-mean normalized least square distance correlation function is rewritten as: In the above formula: X=(x 0 ,y 0 ,z 0 ) T represents the coordinates of the center point of the reference sub-block, ξ=(Δx, Δy, Δz) T is the local coordinate corresponding to the center point, is the initial value of iteration, ΔP=[Δu,Δu x ,Δu y ,Δu w ,Δv,Δv x ,Δv y ,Δv w ,Δw,Δw x ,Δw y ,Δw w ] T is the iterative increment; f(X+W(ξ;ΔP)) and g(X+W(ξ;P 0 )) represent the reference and deformed sub-blocks at X+W(ξ;ΔP) and X+W(ξ;P 0 ) gray value, f m , g m is the average gray value; W(ξ; P 0 ) is expressed as a warping function, which is mainly used to describe the change in the position or shape of the sub-block, while W(ξ;ΔP) is the incremental warping function acting on the reference sub-block. The expressions of the two warping functions are as follows: The first-order iterative increment ΔP is: Among them, Δf is the gray gradient value of each voxel in the volume image; H is the Hessian matrix, which is calculated using the following formula: After the iteration increment ΔP is determined, the incremental warping function W(ξ; ΔP) of the reference sub-block can be determined. Then, the W(ξ; ΔP) in the reference sub-block is compared with the warping function W(ξ; P 0 ) is combined in reverse order, and the new warping function obtained can be used to update the W(ξ; P 0 ), first updated As shown below: The convergence conditions are: When the calculation result meets the preset convergence extreme value condition ||Δp||≤0.001 or the maximum number of iterations is 10, the deformation vector P = [u,u x ,u y ,u w ,v,v x ,v y ,v w ,w,w x ,w y ,w w ] T , and obtain the final displacement (u, v, w). The corresponding total displacement can be obtained by processing the displacement field. The total displacement is: 。 5. The method for detecting internal deformation of vertebrae by a tomography device according to claim 1, Features: The vertebral loading device in step (1) comprises a frame (1), a turbine loading handle (2), a vortex rod (3), a loading rod (4), a sensor (5), a connecting rod (6), a connecting head (7), a support pad (8), and a static strain gauge (9); the turbine loading handle (2) is connected to the loading rod (4) through the vortex rod (3), the sensor (5) is fixed to the lower end of the loading rod (4), the static strain gauge (9) is connected to the sensor (5), the sensor (5) is connected to the static strain gauge (9), the connecting head (7) is connected to the sensor (5) through the connecting rod (6), and the support pad (8) is placed at the lower part of the frame (1); the upper part of the frame (1) is made of steel, and the left part, the right part and the lower part of the frame are made of PC plastic plate.
Citation Information
Patent Citations
Improved masseter segmentation method based on nuclear magnetic resonance image of distance regularized level set
CN106570860A
3D vertebra CT image active contour segmentation method fusing weighted random forest
CN108510507A
Stereo vision three-dimensional displacement measurement method based on synchronous sub-region search
CN109360246A
Digital image measurement method based on feature matching
CN111354033A
Labeling method, device and equipment for lumbar intervertebral disc and storage medium
CN112528991A
