Three-dimensional image dynamic correction evaluation and orthosis-assisted design method and system thereof

Abstract

A three-dimensional image dynamic correction evaluation and orthosis assisted design method, comprising the following steps: obtaining three-dimensional scanning information of a human body, identifying a plurality of two-dimensional images of the human body, calculating pixel values of these two-dimensional images, and synthesizing an original three-dimensional spinal curve. Synthesizing these two-dimensional images of the human body and the three-dimensional scanning information. Generating a deformed body shape of the human body by an image deformation prediction correction body shape method. Predicting parameter values of the force application position, force application direction, and force application magnitude of an orthosis on the human body according to the deformed body shape by a spinal material property and mechanical model prediction method.

Description

Technical Field

[0001] The present invention relates to a method and system for three-dimensional image dynamic correction evaluation and orthosis assisted design. Background Art

[0002] Currently, the production of scoliosis orthoses still generally uses the traditional plaster casting method. However, the winding method and tightness of the bandage will cause casting errors, and the casting time is relatively long. It is difficult for patients to maintain a fixed posture, resulting in casting errors. Therefore, plaster casting is time-consuming and laborious, and it is difficult to ensure the correct symmetry of the trunk in the front, back, left, and right directions. As a result, the scoliosis orthosis cannot obtain the correct production basis, and the subsequent wearing correction function and comfort of the patient are also affected, and further improvement is needed. Summary of the Invention

[0003] The present invention relates to a method and system for three-dimensional image dynamic correction evaluation and orthosis assisted design, which are used to improve the comfort and treatment effect of patients during the current treatment process.

[0004] According to one aspect of the present invention, a method for three-dimensional image dynamic correction evaluation and orthosis assisted design is provided, including the following steps: obtaining three-dimensional scan information of a human body; obtaining a plurality of two-dimensional images of the human body for identification, calculating the pixel values of these two-dimensional images, and synthesizing an original three-dimensional spine curve; synthesizing these two-dimensional images of the human body and the three-dimensional scan information; generating a deformed body shape of the human body by using an image deformation prediction correction body shape method; predicting the parameter values of the force application position, force application direction, and force application magnitude of the orthosis on the human body according to the deformed body shape by using a spine material property and mechanical model prediction method.

[0005] According to one aspect of the present invention, a system for three-dimensional image dynamic correction evaluation and orthosis assisted design is provided, including a three-dimensional scanning device, an operation unit, an image deformation prediction correction body shape unit, and a spine material property and mechanical model prediction unit. The three-dimensional scanning device is used to obtain three-dimensional scan information of a human body. The operation unit is used to obtain a plurality of two-dimensional images of the human body for identification, calculate the pixel values of these two-dimensional images, and synthesize an original three-dimensional spine curve. The operation unit is used to synthesize these two-dimensional images of the human body and the three-dimensional scan information. The image deformation prediction correction body shape unit is used to generate a deformed body shape of the human body. The spine material property and mechanical model prediction unit predicts the parameter values of the force application position, force application direction, and force application magnitude of the orthosis on the human body according to the deformed body shape. Description of the Drawings

[0006] Figure 1A A schematic diagram showing a three-dimensional image dynamic correction evaluation and orthosis assisted design system according to an embodiment of the present invention;

[0007] Figure 1BSchematic diagram showing a three-dimensional image dynamic correction evaluation and orthosis-assisted design method according to an embodiment of the present invention;

[0008] Figure 2 Schematic diagram showing the superposition of the original three-dimensional spinal curve image and the corrected three-dimensional spinal curve image on the external image of the human body according to an embodiment of the present invention;

[0009] Figure 3 Schematic diagram showing the identification of the three-dimensional spinal type of the human body;

[0010] Figure 4 Schematic diagram showing the force application position, force application direction, and force application magnitude of the orthosis on the human body;

[0011] Figures 5A to 5D Schematic diagram showing the calculation of the elastic potential energy, displacement, minimum elastic potential energy, and deformed body shape of the endpoints on the triangular mesh using the image deformation prediction correction body shape method;

[0012] Figure 6 Schematic diagram showing the image deformation prediction correction body shape method according to an embodiment of the present invention;

[0013] Figure 7 Schematic diagram showing the spinal material property and mechanical model estimation method according to an embodiment of the present invention;

[0014] Figure 8 Schematic diagram showing the Young's modulus (E value) of each vertebral body and intervertebral disc; and

[0015] Figures 9A to 9F Schematic diagram showing the calculation of the elastic potential energy, displacement, minimum elastic potential energy, and deformed body shape of the endpoints on the triangular mesh using the image deformation prediction correction body shape method in another embodiment.

[0016] In the above-mentioned drawings, the meanings of the reference numerals are as follows:

[0017] 100: Three-dimensional image dynamic correction evaluation and orthosis-assisted design system

[0018] 101: Three-dimensional scan information

[0019] 102: Depth image

[0020] 103: Two-dimensional image

[0021] 104: External image of the human body

[0022] 105: Original three-dimensional spinal curve

[0023] 106: Corrected three-dimensional spinal curve

[0024] 107: Elastic potential energy

[0025] 108: Displacement

[0026] 109: Minimum elastic potential energy

[0027] 109A: Deformed body shape

[0028] 110: Three-dimensional scanning device

[0029] 111: Force application position

[0030] 112: Force application direction

[0031] 113: Force application magnitude

[0032] 114: Triangular mesh

[0033] 115: Orthosis

[0034] 116: Finite element model

[0035] 130: Operation unit

[0036] 140: Image deformation prediction and correction body shape unit

[0037] 150: Vertebral material property and mechanical model estimation unit

[0038] C: Coronal plane

[0039] S: Sagittal plane

[0040] T: Vertebral rotation

[0041] P: Pressure

[0042] V: Vertebra

[0043] d, e, f, g, h, i, j, k, l, d, rotation e, rotation f, rotation g, rotation h, rotation i': Points

[0044] g1, g2, i2, i3, g1, rotation g2, rotation h1, rotation i1, rotation i2, rotation i3,: Relay points Detailed implementation manner

[0045] Please refer to Figure 1A and 1B wherein Figure 1A FIG. shows a schematic diagram of a three-dimensional image dynamic correction evaluation and orthosis-assisted design system 100 according to an embodiment of the present invention, Figure 1B FIG. shows a schematic diagram of a three-dimensional image dynamic correction evaluation and orthosis-assisted design method according to an embodiment of the present invention.

[0046] According to an embodiment of the present invention, the system 100 includes a three-dimensional scanning device 110, a computing unit 130, an image deformation prediction and corrected body shape unit 140, and a spinal material property and mechanical model estimation unit 150.

[0047] The three-dimensional scanning device 110 is used to generate a plurality of depth images 102, model the depth images 102 to generate three-dimensional scanning information 101 of the human body, and construct an external shape image 104 of the human body. As Figure 1B shown, in step S110, the depth images 102 are modeled to obtain three-dimensional scanning information 101 of the human body. The three-dimensional scanning device 110 is formed by, for example, multiple depth cameras or lidar modules. In addition to setting up a scanning space that completely covers the limb trunk to be scanned through mechanical design and opto-mechatronics integration to scan the external shape of the human body, it is also equipped with computing and modeling software, which can calculate the depth information of the depth images 102 obtained by scanning and superimpose and model the images, so that the three-dimensional scanning device 110 can establish a three-dimensional human model in a very short time.

[0048] The computing unit 130 obtains the two-dimensional image 103 of the X-ray film to generate an original three-dimensional spinal curve 105 and a corrected three-dimensional spinal curve 106. The original three-dimensional spinal curve 105 is, for example, an image generated by superimposing the X-ray film taken by the patient at present, and the corrected three-dimensional spinal curve 106 is, for example, an image generated by superimposing the estimated possible correction curve using the X-ray film taken by the patient. That is to say, the computing unit 130 processes the two-dimensional image 103 of the X-ray film. For example, the computing unit 130 uses text detection to automatically correct the scale numbers and scale inclination of the X-ray film, and uses optical character recognition (OCR) technology to automatically identify the scale numbers in the two-dimensional image 103 and calculate the actual pixel values. As Figure 1B shown, in step S120, a plurality of two-dimensional images 103 of the human body are obtained for identification, the pixel values of these two-dimensional images 1.03 are calculated, and the original three-dimensional spinal curve 105 is synthesized.

[0049] The synthesized original three-dimensional spinal curve 105 can be compared with the data in the database to identify what type of spinal column the original three-dimensional spinal curve 105 is. If it is determined that the spinal column is a type that needs to be corrected, it is determined what type of curvature this spinal column is to plan a correction plan, and a predicted corrected three-dimensional spinal curve 106 is synthesized, and the original three-dimensional spinal curve 105, the corrected three-dimensional spinal curve 106 and the external shape image 104 of the human body scanned by the three-dimensional scanning device 110 are superimposed. As Figure 1B shown, in step S130, these two-dimensional images 103 of the human body and the three-dimensional scanning information 101 are synthesized.

[0050] In addition, the image deformation prediction and correction body shape unit 140 is used to generate the deformed body shape 109A of the human body. When the operation unit 130 determines that the original three-dimensional spinal curve 105 is a spinal type to be corrected, the image deformation prediction and correction body shape unit 140 can perform the deformation of the three-dimensional spinal curve and the body appearance, display the change process and results of the original three-dimensional spinal curve 105, the corrected three-dimensional spinal curve 106 and the body appearance image 104 of the human body, and can correct this deformation model through big data accumulation. As Figure 1B shown, in step S140, the deformed body shape 109A of the human body is generated by the image deformation prediction and correction body shape method.

[0051] In addition, the spinal material property and mechanical model estimation unit 150 is used to predict the parameter values of the force application position 111, the force application direction 112 and the force application magnitude 113 of the orthosis 115 on the human body. After the analyst obtains the deformation model of the human spine to be corrected, the spinal material property and mechanical model estimation unit 150 can evaluate the parameter values of the force application position 111, the force application direction 112 and the force application magnitude 113 of the orthosis according to the biomechanical model and the results of the front and back body changes, so as to obtain the most appropriate force application parameters as the basis for subsequent treatment management and auxiliary orthosis design. As Figure 1B shown, in step S150, the parameter values of the force application position 111, the force application direction 112 and the force application magnitude 113 of the orthosis on the human body are predicted according to the deformed body shape 109A by the spinal material property and mechanical model estimation method.

[0052] Please refer to Figure 2 and Figure 3 wherein Figure 2 FIG. shows a schematic diagram of superimposing the image of the original three-dimensional spinal curve 105 and the image of the corrected three-dimensional spinal curve 106 on the body appearance image 104 of the human body according to an embodiment of the present invention, Figure 3 FIG. shows a schematic diagram for identifying the type of the three-dimensional spinal curve of the human body. Through the above three-dimensional image dynamic correction evaluation and orthosis auxiliary design system 100 and its method, it can help orthopedic and rehabilitation physicians predict the corrected body shape and perform dynamic correction planning evaluation. In one embodiment, the dynamic correction planning may include estimating the correction period and dynamic auxiliary correction design. The correction progress and expected effect evaluation can assist in judging the correction period required by the patient, and can also reasonably calculate the angle that can be improved and the time required for each adjustment by means of the deformation models of the three-dimensional body shapes before and after correction. In addition, the dynamic correction design can calculate the deformed body shape 109A at each stage and the force application position 111, the force application direction 112 and the force application magnitude 113 of the orthosis 115 on the human body according to the deformation models of the three-dimensional body shapes before and after correction, and summarize and classify and analyze 75 deviation modes of the sagittal plane S for dynamic correction, the coronal plane C and the vertebral body rotation T and other three-plane deviation models, so as to estimate the correction range required for each vertebral body and the correction intensity required.

[0053] Please refer to Figure 4 , which shows a schematic diagram of the force application position 111, force application direction 112, and force application magnitude 113 of the orthosis 115 on the human body. When performing subsequent orthosis assisted design, a biomechanical model of the human spine can be analyzed in conjunction with the finite element method, providing greater assistance for three-dimensional spinal curvature correction and orthosis design. For the optimal quantitative information of spinal curvature orthosis design, such as calculating how much pressure P the orthosis 115 exerts to produce a predetermined reduction in the curvature angle, the direction of the force application point, the area size of the gasket, the stress distribution of each vertebral body, etc., can all be evaluated through the biomechanical model to obtain the treatment planning scheme obtained by intelligent three-dimensional image dynamic correction. After numerical calculation, results such as "curvature angle force application magnitude", "intervertebral disc stress", and "sacral stress" can be obtained. This is for physicians and orthosis design engineers to comprehensively evaluate the numerical analysis results to determine the optimal three-dimensional curvature treatment angle for individual patients and to design the spinal curvature parameters for achieving the treatment plan.

[0054] Please refer to Figure 4 , and a finite element model 116 is established based on the spinal curvature of a specific patient. During the process, geometric construction is performed using the medical images of the patient's computed tomography scan. The geometry of the patient's spine is converted into a finite element model 116 for relevant biomechanical analysis. However, if high-quality three-dimensional hexahedral elements are used in the biomechanical model, it will seriously affect the calculation efficiency. To solve these problems, in this embodiment, X-ray films (two-dimensional images) of the human body's coronal plane and sagittal plane are used to reduce the high-quality three-dimensional model to two two-dimensional equivalent finite element models of the coronal plane C and sagittal plane S. This embodiment uses this model to analyze the force application situation during the spinal curvature correction process, and feeds back the analysis results to the orthosis design. The purpose is to simulate and iteratively improve the spinal curvature orthosis design, and the force application position 111, force application magnitude 113, force application direction 112, and the stress status of each vertebral body V required for calculating the treatment deformation amount can be calculated on the two planes of the coronal plane C and sagittal plane S, so as to be used as the quantitative parameter values during orthosis design.

[0055] Please refer to Figures 5A to 5D and Figure 6 , where Figures 5A to 5D shows a schematic diagram of calculating the elastic potential energy 107, displacement 108, minimum elastic potential energy 109, and deformed body shape 109A of the endpoints on the triangular mesh using the method of predicting the corrected body shape by image deformation, Figure 6 shows a schematic diagram of the steps performed by each part of the image deformation prediction correction body shape unit 140 according to an embodiment of the present invention.

[0056] Please refer to Figure 5A, each endpoint (d, e, f) of each triangular mesh 114 of the body exterior is connected to the nearest point (g, h, i) or node on the original three-dimensional spinal curve 105. Assuming that each line (f - e, f - d, f - i) is pulled by a spring and obeys Hooke's law (F(x) = kx), where k is the elastic coefficient and x is the displacement 108. As shown in Figure 6, taking the example of performing each step by the hardware part and / or software part of the image deformation prediction and correction body type unit 140, in step S210, the endpoints of multiple triangular meshes 114 of the body exterior are respectively connected to the nearest points on the original three-dimensional spinal curve 105, and the elastic potential energy 107 of the endpoints on these triangular meshes 114 is calculated respectively.

[0057] Please refer to Figure 5B , during the spinal correction process, when the body exterior approaches the corrected three-dimensional spinal curve 106 from the original three-dimensional spinal curve 105, such that the points (g, h, i) approach the points (j, k, l), the elastic potential energy of each endpoint (d, e, f) of the triangular mesh 114 is calculated respectively (Ep(x) = 1 / 2kx 2 ). Please refer to Figure 5C , when i approaches l, the three lines (f - e, f - d, f - i) with f as the endpoint will move to the three lines (f' - e, f' - d, f' - l) with f' as the endpoint to obtain the minimum elastic potential energy of each endpoint of the triangular mesh 114, so as to generate the displacement between f and f'. As Figure 6 shown, in step S220, when the original three-dimensional spinal curve 105 approaches the corrected three-dimensional spinal curve 106, the elastic potential energy 107 of the endpoints on these triangular meshes 114 is calculated respectively, such that the elastic potential energy 107 reaches the minimum value of the elastic potential energy 109, so as to generate the displacement 108 of these endpoints.

[0058] Please refer to Figure 5D , according to the corresponding body exterior parameter settings, the elastic potential energy 107 of each endpoint of all triangular meshes 114 is calculated by iterative operation, such that the overall elastic potential energy reaches the minimum value, obtaining a new triangular mesh 114' (d', e', f') and generating a deformed body type according to the displacement 108 of each endpoint. As Figure 6 shown, in step S230, according to the body exterior parameters, the elastic potential energy 107 of the endpoints on these triangular meshes 114 is calculated by iterative operation to make the overall potential energy reach the minimum value, obtaining a new triangular mesh 114' and generating a deformed body type 109A according to the displacement 108 of each endpoint. In an embodiment, when the same force is applied to the cervical vertebra, thoracic vertebra and lumbar vertebra of the body, there will be different deformation amounts. The above-mentioned body exterior parameters refer to different deformation parameters (such as Young's modulus) given according to the difficulty degree of calculating deformation based on the mechanical model of the body.

[0059] As described above, through the aboveFigures 5A to 5D and Figure 6 The image deformation prediction and body shape correction method in Figure 6 can accurately predict the displacement 108 of the respective endpoints of the human body to be corrected, and accordingly generate a deformation model of the three-dimensional body shape before and after correction, as the deformed body shape 109A. The above correction does not limit to complete the correction of the spinal curve at one time, and can complete the correction of the spinal curve in multiple stages according to the judgment of the doctor and the orthosis design engineer. That is to say, the three-dimensional spinal curve 106 after correction is formed by multiple three-dimensional spinal curves after correction at different stages.

[0060] Please refer to Figure 7 and 8 , Figure 7 FIG. shows a schematic diagram of the steps executed by each part of the spinal material property and mechanical model estimation unit 150 according to an embodiment of the present invention. Figure 8 FIG. shows a schematic diagram of the Young's modulus (E value) of each vertebral body and intervertebral disc.

[0061] The method for estimating the material properties and biomechanical model of the vertebral body and intervertebral disc is obtained by iteration through biomechanical simulation combined with experimental regression analysis. In one embodiment, the experimental part is, for example, that the patient is imaged by X-ray in the coronal plane under the action of its own gravity when hanging on a horizontal bar, and the displacement generated by each vertebral body and intervertebral disc under the action of its own gravity is used as the objective function. In the biomechanical model, modeling is carried out under the condition that the patient hangs on a horizontal bar under the action of its own gravity, and optimization regression analysis is carried out in combination with the geometric shapes of the relevant vertebral body V and intervertebral disc to evaluate the stiffness of the vertebral body V and intervertebral disc. As Figure 7 described, taking the example of executing each step by the hardware part and / or software part of the spinal material property and mechanical model estimation unit 150, in step S310, when the human body hangs on a horizontal bar under the action of its own gravity, the human spine is imaged by X-ray, and the displacement generated by each vertebral body and intervertebral disc of the human body under the action of its own gravity is used as the objective function, and optimization regression analysis is carried out to evaluate the stiffness of the vertebral body V and intervertebral disc.

[0062] The optimization regression analysis is as follows: The displacement (δ1) of each element node of each vertebral body V and intervertebral disc can be obtained through iterative calculation of the Young's modulus (E value), and the displacement of each node (δ1) is calculated with the displacement of the objective function for error analysis. The condition for the iteration convergence confirmation of the Young's modulus (E value) is that the root mean square value of the displacements of these nodes reaches the minimum value (the error is about 0.1 to 0.3). The Young's modulus (E value) of each vertebral body V and intervertebral disc obtained after calculation is as Figure 8 shown. As Figure 7As described above, in step S320, the iterative calculation of the force application state when the spine is straightened can obtain the displacements of each vertebral body V and the disc element nodes of the human body. Error analysis is performed on the displacements of these nodes and the displacements of the objective function to minimize the root mean square value of these displacements, and the parameter values of the most appropriate force application position 111, force application direction 112, and force application magnitude 113 are obtained.

[0063] Please refer to Figures 9A to 9F , which shows a schematic diagram of calculating the elastic potential energy 107, displacement 108, minimum elastic potential energy 109, and deformed body shape 109A of the endpoints on the triangular mesh 114 by using the image deformation prediction and correction body shape method in another embodiment.

[0064] Please refer to Figure 9A , and according to the relay point density substituted by the Young's modulus, points with corresponding density are generated on the original three-dimensional spine curve 105. In addition, please refer to Figure 9B , and a number of relay points are generated inside the body according to the point cloud density. The relay points can be generated according to the human body structure or evenly generated, for example. Please refer to Figure 9C , each endpoint (d, e, f) of each triangular mesh 114 of the body shape is connected to the nearest relay point (g1, h1, i1) respectively, and possible relay points are searched for. For example: the relay point (g1) is then connected to the nearest point (g) on the original spine curve 105 by the relay point (g2), and the relay point (i1) is connected to the nearest point (i) on the original spine curve 105 by the relay points (i2, i3). Please refer to Figure 9D , assuming that each connected line is pulled by a spring and obeys Hooke's law (F(x) = kx), and the current length is the state of the lowest potential energy. Please refer to Figure 9E , during the spine correction process, when the spine is adjusted or rotated according to the user's needs, and the body shape approaches the corrected three-dimensional spine curve 106 from the original three-dimensional spine curve 105, the nodes (g, h, i) approach the nodes (g', h', i'), so that each relay point (g1, g2, h1, i1, i2, i3) moves to the new anchor points (g1', g2', h1', i1', i2', i3'), and the elastic potential energy 107 of each endpoint and relay point of the triangular mesh 114 is calculated respectively to make the elastic potential energy 107 reach the minimum elastic potential energy 109, so as to generate the displacement and rotation amounts of these endpoints. Please refer to Figure 9E and 9F , according to the body shape parameters, the elastic potential energy 107 of the endpoints and relay points on these triangular meshes 114 is calculated by iterative operation to make the overall elastic potential energy reach the minimum value, and a new triangular mesh 114' (d', e', f') is obtained to generate the deformed body shape 109A.

[0065] As described above, through the above Figures 9A to 9DThe image deformation prediction and correction body shape method in [it] can accurately predict the displacement amount 108 and rotation amount of each end point of the human body to be corrected, and accordingly generate a deformation model of the three-dimensional body shape before and after correction as the deformed body shape.

[0066] The three-dimensional image dynamic correction evaluation and orthosis 115 assisted design method of the above embodiment of the present invention can obtain three-dimensional scan information through three-dimensional scanning, and synthesize the two-dimensional image and three-dimensional scan information of the human body to identify the three-dimensional spine curve type of the human body. In addition to providing designers with accurate three-dimensional spine curves after correction, it can also accumulate big data to provide analysts with the required dynamic correction evaluation and orthosis 115 assisted design, overcoming the disadvantages of making orthosis 115 by traditional plaster casting methods, so as to improve the comfort and treatment effect of current treatment courses for patients.

[0067] The specific embodiments described above have further elaborated on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A three-dimensional image dynamic correction evaluation and orthosis assisted design method, comprising: Obtain three-dimensional scanning information of a human body; Obtain multiple two-dimensional images of the human body for identification, calculate the pixel values of these two-dimensional images, and synthesize an original three-dimensional spinal curve; Synthesize these two-dimensional images of the human body and the three-dimensional scanning information; Generate a deformed body shape of the human body by an image deformation prediction and body shape correction method; and Predict parameter values of the force application position, force application direction, and force application magnitude of an orthosis on the human body based on the deformed body shape by a spinal material property and mechanical model estimation method, wherein the spinal material property and mechanical model estimation method includes: With the human body hanging on a horizontal bar under the action of its own gravity, obtain these two-dimensional images of the human body, and use the displacements generated by each spinal body and intervertebral disc of the human body under the action of its own gravity as the objective function to perform an optimization regression analysis to evaluate the stiffness of the spinal body and intervertebral disc; Through iterative calculation of the force application state when the spine is straightened, the displacements of the element nodes of each vertebral body and intervertebral disc of the human body can be obtained. Perform an error analysis on the displacements of these nodes and the displacements of the objective function to minimize the root mean square value of the displacements of these nodes, and obtain the parameter values of the force application position, the force application direction, and the force application magnitude.

2. The three-dimensional image dynamic correction evaluation and orthosis assisted design method according to claim 1, wherein the image deformation prediction and correction body shape method comprises: Overlay these two-dimensional images of the human body to generate the original three-dimensional spinal curve and the corrected three-dimensional spinal curve, and overlay the original three-dimensional spinal curve and the corrected three-dimensional spinal curve on the external shape image of the human body; Connect the endpoints of multiple triangular meshes of the overlaid external shape image of the human body to the nearest points on the original three-dimensional spinal curve respectively, and calculate the elastic potential energy of the endpoints on these triangular meshes respectively; When the original three-dimensional spinal curve approaches the corrected three-dimensional spinal curve, calculate the elastic potential energy of the endpoints on these triangular meshes to minimize the elastic potential energy, so as to generate the displacement amounts of these endpoints; and According to the body shape parameters, perform iterative calculations to calculate the elastic potential energy of the endpoints on these triangular meshes to minimize the overall elastic potential energy, obtain new triangular meshes, and generate the deformed body shape according to the displacement amounts of their respective endpoints.

3. The three-dimensional image dynamic correction evaluation and orthosis assisted design method according to claim 1, wherein the image deformation prediction and correction body shape method comprises: Overlay these two-dimensional images of the human body to generate the original three-dimensional spinal curve and the corrected three-dimensional spinal curve, and overlay the original three-dimensional spinal curve and the corrected three-dimensional spinal curve on the external shape image of the human body; Establish multiple relay points inside the human body, connect the endpoints of multiple triangular meshes of the body shape to the nearest relay points respectively, and connect these relay points to the nearest points on the original three-dimensional spinal curve respectively; Calculate the elastic potential energy of the endpoints on these triangular meshes and these relay points respectively; When the original three-dimensional spinal curve approaches the corrected three-dimensional spinal curve, calculate the elastic potential energy of the endpoints on these triangular meshes and these relay points to minimize the elastic potential energy, so as to generate the displacement amounts and rotation amounts of these endpoints; and According to the body shape parameters, perform iterative calculations to calculate the elastic potential energy of the endpoints on these triangular meshes and these relay points to minimize the overall elastic potential energy, obtain new triangular meshes, and generate the deformed body shape according to the displacement amounts of their respective endpoints.

4. The three-dimensional image dynamic correction evaluation and orthosis assisted design method according to claim 1, wherein the optimization regression analysis includes obtaining the displacements of each vertebral body and disc element node through iterative calculation of Young's modulus, and the condition for the iterative convergence confirmation of Young's modulus is that the root mean square value of the displacements of these nodes reaches the minimum value.

5. The three-dimensional image dynamic correction evaluation and orthosis assisted design method according to claim 1, wherein obtaining and identifying these two-dimensional images of the human body includes automatically identifying these two-dimensional images using text detection or optical character recognition.

6. The three-dimensional image dynamic correction evaluation and orthosis assisted design method according to claim 1, wherein the spinal material property and mechanical model prediction method includes analyzing the biomechanical model of the human spine by the finite element method, and reducing the three-dimensional model to two two-dimensional equivalent finite element models of the coronal plane and sagittal plane of the human body using these two-dimensional images of the coronal plane and sagittal plane of the human body.

7. The three-dimensional image dynamic correction evaluation and orthosis assisted design method as claimed in claim 1, wherein the dynamic correction evaluation includes calculating the deformed body shapes at each stage, the force application positions, the force application directions, and the force application magnitudes based on the deformation models of the three-dimensional body shapes before and after correction.

8. The three-dimensional image dynamic correction evaluation and orthosis assisted design method as claimed in claim 1, wherein the orthosis assisted design includes calculating the pressure exerted by the orthosis to be able to produce a reduction in the predetermined scoliosis angle, the direction of the force application point, the area size of the spacer, and the stress distribution of each vertebral body based on the deformation models of the three-dimensional body shapes before and after correction.

9. The three-dimensional image dynamic correction evaluation and orthosis assisted design method as claimed in claim 2, wherein the corrected three-dimensional spinal curve is a plurality of corrected three-dimensional spinal curves at different stages.

10. A three-dimensional image dynamic correction evaluation and orthosis assisted design system, comprising: A three-dimensional scanning device for obtaining three-dimensional scanning information of a human body; An operation unit is used to obtain multiple two-dimensional images of the human body for identification, calculate the pixel values of these two-dimensional images, and synthesize an original three-dimensional spinal curve. The operation unit is used to synthesize these two-dimensional images of the human body and the three-dimensional scanning information; An image deformation prediction and body shape correction unit is used to generate the deformed body shape of the human body; And A spinal material property and mechanical model estimation unit predicts the parameter values of the force application position, force application direction, and force application magnitude of the orthosis on the human body according to the deformed body shape, wherein the spinal material property and mechanical model estimation unit includes: The first part obtains these two-dimensional images of the human body when the human body hangs on a horizontal bar under the action of its own gravity. The displacements generated by each spinal body and intervertebral disc of the human body under the action of its own gravity are used as the objective function, and an optimization regression analysis is performed to evaluate the stiffness of the spinal body and intervertebral disc; And The second part can obtain the displacements of the element nodes of each vertebral body and intervertebral disc of the human body through iterative calculations of the force application state when the spine is straightened. An error analysis is performed on the displacements of these nodes and the displacements of the objective function to minimize the root mean square value of the displacements of these nodes, and the parameter values of the force application position, force application direction, and force application magnitude are obtained.

11. The three-dimensional image dynamic correction evaluation and orthosis assisted design system as claimed in claim 10, wherein the image deformation prediction and corrected body shape unit includes: The first part is used to superimpose these two-dimensional images of the human body to generate the original three-dimensional spinal curve and the corrected three-dimensional spinal curve, and superimpose the original three-dimensional spinal curve and the corrected three-dimensional spinal curve on the external shape image of the human body; The second part is used to connect the endpoints of multiple triangular meshes of the superimposed external shape image of the human body to the nearest points on the original three-dimensional spinal curve respectively, and calculate the elastic potential energy of the endpoints on these triangular meshes respectively; When the original three-dimensional spinal curve approaches the corrected three-dimensional spinal curve, the second part calculates the elastic potential energy of the endpoints on these triangular meshes to minimize the elastic potential energy, so as to generate the displacement amounts of these endpoints; And The third part, according to the body shape parameters, calculates the elastic potential energy of the endpoints on these triangular meshes through iterative operations to minimize the overall elastic potential energy, obtains new triangular meshes, and generates the deformed body shape according to the displacement amounts of their respective endpoints.

12. The three-dimensional image dynamic correction evaluation and orthosis assisted design system as claimed in claim 10, wherein the image deformation prediction and corrected body shape unit includes: The first part is used to superimpose these two-dimensional images of the human body to generate the original three-dimensional spinal curve and the corrected three-dimensional spinal curve, and superimpose the original three-dimensional spinal curve and the corrected three-dimensional spinal curve on the external shape image of the human body; The second part is used to establish multiple relay points inside the human body, connect the endpoints of the complex triangular meshes of the body shape to the nearest relay points respectively, and connect these relay points to the nearest points on the original three-dimensional spinal curve respectively; The third part is used to calculate the elastic potential energy of the endpoints on these triangular meshes and these relay points respectively. When the original three-dimensional spinal curve approaches the corrected three-dimensional spinal curve, the third part calculates the elastic potential energy of the endpoints on these triangular meshes and these relay points to minimize the elastic potential energy, so as to generate the displacement amounts and rotation amounts of these endpoints; And In the fourth part, based on the body shape parameters, iteratively calculate the elastic potential energy of the endpoints and the relay points on these triangular meshes to minimize the overall elastic potential energy, obtain a new triangular mesh, and generate the deformed body shape according to the displacement of each endpoint.

13. The three-dimensional image dynamic correction evaluation and orthosis assisted design system according to claim 10, wherein the optimization regression analysis includes obtaining the displacements of each vertebral body and disc element node through iterative calculation of Young's modulus, and the condition for confirming the convergence of the Young's modulus iteration is that the root mean square value of the displacements of these nodes reaches the minimum value.

14. The three-dimensional image dynamic correction evaluation and orthosis assisted design system according to claim 10, wherein the spinal material property and mechanical model prediction unit analyzes the biomechanical model of the human spine by the finite element method, and reduces the three-dimensional model to two two-dimensional equivalent finite element models of the coronal plane and the sagittal plane by using these two-dimensional images of the coronal plane and the sagittal plane of the human body.

15. A three-dimensional image dynamic correction evaluation and orthosis assisted design method, comprising: Obtain the three-dimensional scan information of the human body; Obtain multiple two-dimensional images of the human body for identification, calculate the pixel values of these two-dimensional images, and synthesize the original three-dimensional spine curve; Synthesize these two-dimensional images of the human body and the three-dimensional scan information; Generate the deformed body shape of the human body by the method of predicting and correcting the body shape through image deformation; and Predict the parameter values of the force application position, force application direction, and force application magnitude of the orthosis on the human body according to the deformed body shape by the method of estimating based on spine material characteristics and mechanical models, where the method of predicting and correcting the body shape through image deformation includes: Overlay these two-dimensional images of the human body to generate the original three-dimensional spine curve and the corrected three-dimensional spine curve, and overlay the original three-dimensional spine curve and the corrected three-dimensional spine curve on the body shape image; Connect the endpoints of multiple triangular meshes of the overlaid body shape image to the nearest points on the original three-dimensional spine curve respectively, and calculate the elastic potential energy of the endpoints on these triangular meshes respectively; When the original three-dimensional spine curve approaches the corrected three-dimensional spine curve, calculate the elastic potential energy of the endpoints on these triangular meshes to minimize the elastic potential energy, so as to generate the displacement of these endpoints; and Based on the body shape parameters, iteratively calculate the elastic potential energy of the endpoints on these triangular meshes to minimize the overall elastic potential energy, obtain a new triangular mesh, and generate the deformed body shape according to the displacement of each endpoint.

16. A three-dimensional image dynamic correction evaluation and orthosis assisted design method, comprising: Obtain the three-dimensional scan information of the human body; Obtain multiple two-dimensional images of the human body for identification, calculate the pixel values of these two-dimensional images, and synthesize the original three-dimensional spine curve; Synthesize these two-dimensional images of the human body and the three-dimensional scan information; Generate the deformed body shape of the human body by the method of predicting and correcting the body shape through image deformation; and Predict the parameter values of the force application position, force application direction, and force application magnitude of the orthosis on the human body according to the deformed body shape by the method of estimating based on spine material characteristics and mechanical models, where the method of predicting and correcting the body shape through image deformation includes: Overlay these two-dimensional images of the human body to generate the original three-dimensional spine curve and the corrected three-dimensional spine curve, and overlay the original three-dimensional spine curve and the corrected three-dimensional spine curve on the body shape image; Establish multiple relay points inside the human body, connect the endpoints of multiple triangular meshes of the body shape to the nearest relay points respectively, and connect these relay points to the nearest points on the original three-dimensional spine curve respectively; Calculate the elastic potential energy of the endpoints and the relay points on these triangular meshes respectively; When the original three-dimensional spine curve approaches the corrected three-dimensional spine curve, calculate the elastic potential energy of the endpoints and the relay points on these triangular meshes to minimize the elastic potential energy, so as to generate the displacement and rotation of these endpoints; and Based on the body shape parameters, iterative calculations are performed to calculate the elastic potential energy of the endpoints and the relay points on these triangular meshes so that the overall elastic potential energy reaches the minimum value of the elastic potential energy, obtaining a new triangular mesh and generating the deformed body shape according to the displacement amounts of their respective endpoints.

Images