Pedicle screw placement path three-dimensional visualization system for spinal trauma

Abstract

The present application relates to the field of medical image technology, in particular to a kind of for spinal trauma's pedicle screw placement path three-dimensional visualization system.Data acquisition module obtains the three-dimensional CT image of patient spine, and bone tissue voxel is filtered according to the voxel block CT value;Geometric field analysis module determines seed bone region according to bone tissue voxel CT value and position, provides reasonable starting point for anisotropic propagation analysis, and obtains propagation distance index based on CT value;Connectedness analysis module determines connectedness index by analyzing the difference of bone tissue voxel CT value and the similarity of change, determines bone quality connectedness by iterative calculation, avoids free bone block and unstable fracture area;Path planning module calculates the screw placement suitability score based on bone quality connectedness and propagation distance index, filters effective channel voxel, determines screw placement path in combination with its position distribution and score, and plans the optimal path that fits the individualized spine structure of patient.

Description

Technical Field

[0001] This invention relates to the field of medical imaging technology, and more specifically to a three-dimensional visualization system for pedicle screw placement paths in spinal trauma. Background Technology

[0002] A burst fracture of the spine is a severe spinal injury characterized by the bursting of the vertebral body under vertical compressive force. Fracture fragments may displace in all directions or even protrude into the spinal canal, compressing nerves. Pedicle screw fixation is the primary treatment for this type of fracture. The key to surgical success lies in accurately placing the screw within the narrow bony passage of the pedicle and ensuring that the screw is firmly anchored to stable bone.

[0003] Current clinical surgical planning primarily relies on the Hounsfield Unit (CT) values ​​from CT images to identify bone tissue. The conventional approach is to set a global CT threshold (e.g., 200 HU), treating areas above this threshold as bone and calculating their geometric center as the recommended path. However, in burst fractures, the resulting free fragments, while still appearing as high grayscale values ​​on CT images, have broken from the vertebral body's main structure and lost their mechanical support. Grayscale threshold-based algorithms cannot distinguish between high-density effective bone wall and high-density ineffective free fragments. If the path planning algorithm fails to identify and avoid these free fragments, the calculated path may guide the screw through the unstable fracture area, leading to screw loosening, pullout, or internal fixation failure, thus failing to achieve the desired treatment outcome. Summary of the Invention

[0004] To address the technical problem that grayscale threshold-based algorithms cannot distinguish between high-density effective bone wall and high-density ineffective free bone fragments, and that if the path planning algorithm fails to identify and avoid these free bone fragments, the calculated path may guide the screw through an unstable fracture area, the present invention aims to provide a three-dimensional visualization system for pedicle screw placement paths in spinal trauma. The specific technical solution adopted is as follows: This invention proposes a three-dimensional visualization system for pedicle screw placement paths in spinal trauma, the system comprising: The data acquisition module is used to acquire three-dimensional CT images of the patient's spine; in the three-dimensional CT images, bone tissue voxels are selected based on the numerical characteristics of the CT values ​​of the voxel blocks. The geometric field analysis module is used to determine the seed bone region based on the CT value and location characteristics of bone tissue voxels and to perform anisotropic propagation based on the CT value of bone tissue voxels, thereby obtaining the propagation distance index of each bone tissue voxel. The connectivity analysis module is used to analyze the differences in CT values ​​and the similarity in changes of CT values ​​between bone tissue voxels to determine the connectivity index between bone tissue voxels. Based on the seed bone region, iterative calculations are performed based on the connectivity index between bone tissue voxels to determine the bone connectivity of each bone tissue voxel. The path planning module is used to calculate the pin placement suitability score for each bone tissue voxel based on the bone connectivity of each bone tissue voxel and to use the propagation distance index for spatial constraints, and to screen effective channel voxels; based on the location distribution characteristics of effective channel voxels and the pin placement suitability score, the pin placement path is determined.

[0005] Furthermore, the method for obtaining the seed bone region includes: Connectivity analysis was performed on all bone tissue voxels to obtain all connected components of the bone tissue. In each bone tissue connected region, the sum of the CT values ​​of all bone tissue voxels is calculated as the CT feature value of each bone tissue connected region; Among all bone tissue connected regions, the bone tissue connected region with the largest CT feature value is selected as the seed bone region.

[0006] Furthermore, the method for obtaining the propagation distance index includes: The propagation distance index of bone tissue voxels in the seed bone region is set to a first preset value, and the propagation distance index of bone tissue voxels in the non-seed bone region is set to a second preset value, wherein the first preset value is less than the second preset value. For each bone tissue voxel in the non-seed bone region, the sum of its CT value and a preset constant is used as the local propagation velocity. Starting from the bone tissue voxel in the seed bone region, the anisotropic propagation equation is solved using the fast travel algorithm to calculate the minimum cumulative propagation cost for the bone tissue voxel in the non-seed bone region to reach the bone tissue voxel in the seed bone region. This cost replaces the second preset value as a new propagation distance index for the bone tissue voxel in the non-seed bone region.

[0007] Furthermore, the method for obtaining the connectivity index includes: Determine the local neighborhood of each bone tissue voxel centered on each bone tissue voxel; Within the local neighborhood of each bone tissue voxel, the absolute value of the difference between the CT values ​​of the central voxel and each neighboring voxel is negatively correlated and normalized to obtain the similarity factor between the central voxel and each neighboring voxel. Calculate the CT value gradient modulus for each bone tissue voxel; Within the local neighborhood of each bone tissue voxel, the maximum value of the CT value gradient modulus between the central voxel and each neighboring voxel is negatively correlated and normalized, and this value is used as the connectivity factor between the central voxel and each neighboring voxel. The normalized value of the product of the similarity factor and the connectivity factor between the central voxel and each neighboring voxel is used as the connectivity index between each central voxel and each neighboring voxel.

[0008] Furthermore, the method for obtaining bone connectivity includes: The bone connectivity of each bone tissue voxel is initialized, and the initial values ​​of bone connectivity of bone tissue voxels in the seed bone region and the non-seed bone region are different. In each iteration, for any bone tissue voxel, the formulaic model for the bone connectivity of that voxel in that iteration includes: in, This indicates the bone connectivity of the bone tissue voxel in this iteration; Indicates the damping factor; This represents the initial value of bone connectivity for the voxel in the bone tissue. This indicates the number of neighboring voxels of the bone tissue voxel; This represents the bone connectivity of the i-th neighboring voxel of the bone tissue in the previous adjacent iteration; This represents the connectivity index between the bone tissue voxel and its i-th neighboring voxel; After each iteration, the bone connectivity of all bone tissue voxels is used to form a connectivity numerical vector. The iteration stops when the Euclidean distance between the connectivity numerical vectors of two adjacent iterations is less than a preset convergence threshold, and the bone connectivity of each bone tissue voxel in the last iteration is obtained.

[0009] Furthermore, the method for obtaining the nail placement suitability score includes: When the propagation distance index of a certain bone tissue voxel is less than the preset distance threshold, the distance cutoff value of the bone tissue voxel is set to 1; otherwise, it is set to 0. The connectivity enhancement factor is obtained by exponentially calculating the bone connectivity value of each bone tissue voxel. The connectivity enhancement factor of each bone tissue voxel is multiplied by the distance cutoff value to obtain the pin placement suitability score for each bone tissue voxel.

[0010] Furthermore, the method for obtaining the effective channel voxel includes: All bone tissue voxels were sorted in ascending order for nail placement suitability scores, and the nail placement suitability scores at the preset percentile threshold were used as the score screening threshold. Bone tissue voxels with a pin placement suitability score greater than or equal to a preset score screening threshold are considered as valid channel voxels.

[0011] Furthermore, the method for obtaining the pin placement path includes: Obtain the geometric center of the seed bone region and construct a sagittal segmentation plane, wherein the sagittal segmentation plane passes through the geometric center and its normal vector is parallel to the horizontal direction; The effective channel voxels are divided into two subsets using the sagittal segmentation plane; In each subset, the coordinates of the effective channel voxels in the 3D CT image are weighted and averaged using the pin placement suitability score of the effective channel voxels to obtain the weighting center of each subset. A covariance matrix is ​​constructed based on the three coordinate axes in three-dimensional space. At each element position, a weighted average is performed using the weighting center of each subset, the pinning suitability score of the effective channel voxel in each subset, and the coordinates of the effective channel voxel to obtain the value of each element in the covariance matrix. The covariance matrix corresponding to each subset is decomposed into eigenvalues, and the eigenvector corresponding to the largest eigenvalue is used as the nailing direction in each subset. A local cylindrical coordinate system is established based on the nail-feeding direction. The vertical distance distribution from the effective channel voxel set to the straight line containing the nail-feeding direction is statistically analyzed, and the maximum safe channel radius of the effective channel voxels in each subset is determined.

[0012] Furthermore, the method for obtaining the maximum safe passage radius includes: Arrange the vertical distances between all valid channel voxels in each subset and the straight line containing the nail insertion direction in ascending order to obtain a distance sequence; In the distance sequence, the vertical distance at the preset percentile is used as the maximum safe passage radius.

[0013] Furthermore, the method for obtaining the bone tissue voxel includes: In the three-dimensional CT image, voxel blocks with CT values ​​greater than a preset CT threshold are taken as bone tissue voxels, wherein the CT value in the three-dimensional CT image is a normalized value, and the preset CT threshold ranges from [0.3, 0.5].

[0014] The present invention has the following beneficial effects: The data acquisition module acquires 3D CT images of the patient's spine, providing a comprehensive and multi-angle view of the spinal anatomy. Based on the numerical characteristics of the CT values ​​of voxel blocks, bone tissue voxels are screened, enabling rapid extraction of bone tissue-related information from complex 3D image data. Then, in the geometric field analysis module, seed bone regions are determined based on the CT values ​​and location characteristics of the bone tissue voxels. This allows for more accurate localization of representative and critical bone tissue areas, providing a reasonable starting point for subsequent anisotropic propagation analysis. Anisotropic propagation based on the CT values ​​of bone tissue voxels yields the propagation distance index for each voxel, effectively preventing deviations from the correct anatomical range during subsequent path planning. Furthermore, to understand the microscopic continuity of bone structure, the connectivity analysis module analyzes the differences in CT values ​​and the similarity of CT value changes between bone tissue voxels to determine the connectivity index between any two voxels, reflecting the continuity and integrity of the bone tissue. Thus, based on the seed bone regions, iterative calculations using the connectivity index between bone tissue voxels determine the bone connectivity of each voxel, avoiding free bone fragments and ensuring that subsequent paths avoid unstable fracture areas. Finally, in the path planning module, based on the bone connectivity of each bone tissue voxel and using the propagation distance index for spatial constraints, the screw placement suitability score of each bone tissue voxel is calculated. Effective channel voxels are selected based on the screw placement suitability score, and the screw placement path is determined according to the location distribution characteristics of the effective channel voxels and the screw placement suitability score. This fully utilizes the advantages of effective channel voxels and, combined with the patient's individualized spinal structural characteristics, plans the optimal screw placement path. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a system block diagram of a three-dimensional visualization system for pedicle screw placement path in spinal trauma, provided in one embodiment of the present invention. Figure 2 This is a schematic diagram of the system structure of a three-dimensional visualization system for pedicle screw placement path in spinal trauma, provided by an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a three-dimensional visualization system for pedicle screw placement in spinal trauma proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for a three-dimensional visualization system for pedicle screw placement in spinal trauma provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a system block diagram of a three-dimensional visualization system for pedicle screw placement path in spinal trauma according to an embodiment of the present invention. The system includes: a data acquisition module 101, a geometric field analysis module 102, a connectivity analysis module 103, and a path planning module 104.

[0021] The data acquisition module 101 is used to acquire three-dimensional CT images of the patient's spine; in the three-dimensional CT images, bone tissue voxels are selected based on the numerical characteristics of the CT values ​​of the voxel blocks.

[0022] To obtain high-quality 3D CT images, scanning parameters can be precisely set according to the patient's specific situation (such as age, body type, and degree of spinal lesions). Then, a computed tomography (CT) scanner is used to scan the patient's spine. After scanning, the CT equipment transmits the acquired raw data to the hospital's information management system (HIS) via a high-speed data interface (such as a DICOM interface), thus obtaining a 3D CT image of the patient's spine. Simultaneously, to eliminate the influence of different patients, different scanning equipment, or different scanning parameters on the images, the CT values ​​in the images need to be normalized. Specifically, an upper limit of 1250 HU and a lower limit of -250 HU can be set to normalize the CT value of each voxel in the CT image. The normalization formula is as follows: ;in, The normalized CT value. The original CT values. This is the lower limit value. The value is the upper limit. If the calculation result is less than 0, it is forced to be 0. If the calculation result is greater than 1, it is forced to be 1. This way, the normalized CT value of all voxels in the three-dimensional CT image can be obtained.

[0023] To reduce computational load and eliminate soft tissue interference, it is necessary to extract regions containing only bone tissue. Since bone tissue usually exhibits a high CT value in CT images, bone tissue voxels can be selected based on the numerical characteristics of the CT value of voxel blocks in 3D CT images.

[0024] Preferably, in one embodiment of the present invention, the method for obtaining bone tissue voxels includes: In 3D CT images, voxel blocks with CT values ​​greater than a preset CT threshold are taken as bone tissue voxels. Since the CT value in 3D CT images is a normalized value, the preset CT threshold ranges from [0.3, 0.5]. In this embodiment of the present invention, it is set to 0.3. The specific value and range can be adjusted according to the implementation scenario and are not limited here.

[0025] In the embodiments of the present invention, the collection and acquisition of personal information data are authorized by the relevant users, and the process does not violate relevant laws and regulations, nor does it violate public order and good morals.

[0026] The geometric field analysis module 102 is used to determine the seed bone region based on the CT value and location characteristics of the bone tissue voxels and to perform anisotropic propagation based on the CT value of the bone tissue voxels, thereby obtaining the propagation distance index of each bone tissue voxel.

[0027] The main function of this module is to provide a reference point for calculation and introduce geometric constraints when determining the subsequent path. Therefore, it is necessary to find an absolutely safe core area in the vertebral body of the patient's spine as the seed bone area, and use this as the starting point to calculate the spatial propagation distance to prevent the subsequent path planning from deviating from the anatomical range.

[0028] CT value is an important indicator for measuring bone tissue density. In three-dimensional CT images of the spine, the CT values ​​of bone tissue in different locations will vary due to differences in composition and structure. At the same time, the positional characteristics of bone tissue in the spine reflect its anatomical relationship and functional role. Different bone tissues have specific positions and arrangements in the spine, and they are interconnected and influence each other. Therefore, considering the CT value characteristics of bone tissue voxels while combining them with positional characteristics can provide a more comprehensive consideration of the spatial relationship of bone tissue and avoid the possible deviations that may occur when determining the sesamoid bone region based solely on CT values.

[0029] Preferably, in one embodiment of the present invention, the method for obtaining the seed bone region includes: Bone tissue typically exists as a continuous structure in the spine, so connected component analysis is performed on all bone tissue voxels to obtain all bone tissue connected components.

[0030] Because burst fractures cause vertebral body fragmentation, these ineffective free bone fragments often exhibit the same CT values ​​as normal bone fragments, but are usually smaller in area. Therefore, within each connected bone tissue region, the sum of the CT values ​​of all bone tissue voxels is calculated as the CT feature value for each region. A larger CT feature value indicates a greater number of bone tissue voxels within the region, and a higher CT value also suggests a higher likelihood of it being a seed bone region. Therefore, among all connected bone tissue regions, the region with the largest CT feature value is selected as the seed bone region.

[0031] It should be noted that connected component analysis is a well-known technique, and the specific process will not be elaborated here.

[0032] Anisotropic propagation is a propagation method that considers directionality. Propagation based on the CT values ​​of bone tissue voxels can simulate the growth, diffusion, or mechanical conduction of bone tissue under physiological or pathological conditions. The propagation speed and range in different directions vary depending on the density of the bone tissue (reflected by the CT values) and the characteristics of the surrounding tissues, which is more realistic. The propagation distance index of each bone tissue voxel obtained through anisotropic propagation can quantitatively describe the spatial relationship and propagation characteristics between the bone tissue voxel and the seed bone region. This index can provide important geometric constraints for subsequent path analysis.

[0033] Preferably, in one embodiment of the present invention, the method for obtaining the propagation distance index includes: The propagation distance index of bone tissue voxels within the seed bone region is set as a first preset value, and the propagation distance index of bone tissue voxels outside the seed bone region is set as a second preset value, wherein the first preset value is less than the second preset value. In this embodiment of the invention, the first preset value is set to 0, and the second preset value is set to 10. 6 , is used to represent infinity.

[0034] Before calculating the propagation distance index, the allowed propagation velocity is first defined: for each bone tissue voxel in a non-seed bone region, the sum of its CT value and a preset constant is used as the local propagation velocity. Here, the preset constant is set to 0.01 to prevent the calculation from stalling due to a velocity of 0. The denser the bone, that is, the larger the CT value, the faster the wavefront propagation and the smaller the distance cost. Conversely, the slower the propagation and the greater the distance cost, the less dense the bone.

[0035] Then, starting from the bone tissue voxel in the seed bone region, the anisotropic propagation equation is solved using the fast travel algorithm to obtain the minimum cumulative propagation cost for the bone tissue voxel in the non-seed bone region to reach the bone tissue voxel in the seed bone region. In this embodiment of the invention, the minimum cumulative propagation cost is used to replace the second preset value as a new propagation distance index for the bone tissue voxel in the non-seed bone region. The smaller the value, the closer the bone tissue voxel in the non-seed bone region is to the seed bone region (vertebral core) and the better the bone quality along the way; conversely, the larger the value, the farther the bone tissue voxel in the non-seed bone region is from the seed bone region (vertebral core).

[0036] It should be noted that the fast-moving algorithm is a well-known technique, and the specific process will not be elaborated here.

[0037] The connectivity analysis module 103 is used to analyze the differences in CT values ​​and the similarity in changes of CT values ​​between bone tissue voxels, and to determine the connectivity index between bone tissue voxels. Based on the seed bone region, iterative calculations are performed based on the connectivity index between bone tissue voxels to determine the bone connectivity of each bone tissue voxel.

[0038] Bone tissue is a complex three-dimensional structure, and its structural characteristics cannot be fully and accurately assessed using only a single CT value or other local indicators. However, the connectivity analysis module can calculate the connectivity index between two bone tissue voxels by analyzing the differences and similarities in CT values ​​between them, thus providing a preliminary reflection of the local connectivity of the bone tissue.

[0039] Preferably, in one embodiment of the present invention, the method for obtaining the connectivity index includes: Bone tissue is a complex three-dimensional structure, and the voxel properties of different regions may vary greatly. Therefore, we first determine the local neighborhood of each bone tissue voxel with each voxel as the center to limit the scope of the analysis.

[0040] If the CT values ​​of two bone tissue voxels are close and there is no obvious gradient change, it means that they are likely to belong to the same continuous bone. Conversely, if there is a large difference or change, there may be a fracture line.

[0041] Therefore, within the local neighborhood of each bone tissue voxel, the absolute value of the difference between the CT values ​​of the central voxel and each neighboring voxel is calculated. The smaller the absolute value of this difference, the higher the similarity, and the more likely they belong to the same continuous bone tissue. Therefore, this absolute value of the difference is negatively correlated and normalized to correct the logical relationship, thus obtaining the similarity factor between the central voxel and each neighboring voxel. This negative correlation mapping and normalization can be performed using the formula... ,in, Let x represent an exponential function with the natural constant e as the base, and let x represent the independent variable.

[0042] Then, the CT value gradient modulus of each bone tissue voxel is calculated. Within the local neighborhood of each bone tissue voxel, the maximum value of the CT value gradient modulus between the central voxel and each neighboring voxel is selected. If the maximum value is small, it indicates that there is no obvious gradient abrupt change, and the probability that the tissue belongs to the same continuous bone mass is higher. Therefore, the maximum value of the CT value gradient modulus is negatively correlated and normalized to correct the logical relationship, thus obtaining the connectivity factor between the central voxel and each neighboring voxel. The negative correlation mapping and normalization process here can be performed using the formula... ,in, Let x represent an exponential function with the natural constant e as the base, and let x represent the independent variable.

[0043] Finally, the normalized value of the product of the similarity factor and the connectivity factor between the central voxel and each neighboring voxel is used as the connectivity index between each central voxel and each neighboring voxel. Based on the previous analysis, a larger connectivity index indicates better connectivity between the central voxel and its neighboring voxels, and a greater likelihood that they belong to the same continuous bone tissue. The normalization here can be achieved by dividing by the sum of the products of the similarity factor and the connectivity factor between the central voxel and all its neighboring voxels.

[0044] It should be noted that in this embodiment of the invention, the size of the local neighborhood is set to 26 neighborhoods. The specific size can be adjusted according to the implementation scenario and is not limited here.

[0045] The connectivity between local bone tissue voxels affects the mechanical properties and stability of the entire bone tissue. Therefore, starting from the seed bone region, iterative calculations based on the connectivity index between bone tissue voxels can gradually consider the interaction between bone tissue voxels, extend the local connectivity to the whole, determine the bone connectivity of each bone tissue voxel, and thus more accurately reflect the actual connectivity of the bone tissue.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining bone connectivity includes: The bone connectivity of each bone tissue voxel is initialized, and the initial values ​​of bone connectivity of bone tissue voxels in the seed bone region and bone tissue voxels in the non-seed bone region are different. In this embodiment of the invention, the initial value of bone connectivity of bone tissue voxels in the seed bone region is set to a non-zero value, such as the reciprocal of the number of bone tissue voxels, which represents an absolutely reliable vertebral core and the source of power. The initial value of bone connectivity of bone tissue voxels in the non-seed bone region is set to 0, which represents the absence of energy, thereby forming a "potential energy difference".

[0047] To identify free bone fragments, a random walk model from graph theory can be used to measure the topological affinity between each bone tissue voxel and the seed bone region, essentially establishing an iterative process: In each iteration, for any given bone tissue voxel, the formula for the bone connectivity of that voxel in that iteration includes: in, This indicates the bone connectivity of the bone tissue voxel in this iteration; Indicates the damping factor; This represents the initial value of bone connectivity for the voxel in the bone tissue. This indicates the number of neighboring voxels of the bone tissue voxel; This represents the bone connectivity of the i-th neighboring voxel of the bone tissue in the previous adjacent iteration; This represents the connectivity index between the bone tissue voxel and the i-th neighboring voxel.

[0048] In the above formula model, This is to ensure a continuous supply of energy to the seed bone region, meaning that each iteration has... The probability values ​​are directly reset from the seed bone region, ensuring that the numerical flow does not spread indefinitely, but forms a stable distribution field around the seed bone region. This indicates that the energy in the current iteration of each bone tissue voxel is the sum of the energy transmitted from all neighboring voxels. The larger the value, the more complete and high-weighted pathways there are extending from the seed bone region, so that the values ​​can be transmitted continuously. Conversely, for free bone blocks, due to the small connectivity index, the connection lines of the surrounding bone tissue voxels are cut off, so the numerical flow cannot be effectively reached or retained, and the value will be smaller.

[0049] After each iteration, the bone connectivity of all bone tissue voxels is used to form a connectivity vector. The iteration stops when the Euclidean distance between the connectivity vectors of two adjacent iterations is less than a preset convergence threshold, which is considered to have reached a stable equilibrium state. At this time, the bone connectivity of each bone tissue voxel in the last iteration is obtained as the final bone connectivity.

[0050] It should be noted that the damping factor is set to 0.85, and the preset convergence threshold is set to 10. -4 The specific values ​​can be adjusted according to the implementation scenario and are not limited here; in other embodiments of the present invention, the condition for stopping the iteration can also be set to the number of iterations, for example, the bone connectivity of each bone tissue voxel in the 50th iteration can be taken as the final bone connectivity.

[0051] The path planning module 104 is used to calculate the pin placement suitability score of each bone tissue voxel based on the bone connectivity of each bone tissue voxel and to use the propagation distance index for spatial constraints, and to screen effective channel voxels; based on the location distribution characteristics of the effective channel voxels and the pin placement suitability score, the pin placement path is determined.

[0052] Bone connectivity of bone voxels can be used to measure the tightness of connections between bone voxels, reflecting the structural integrity and mechanical properties of bone tissue. During screw placement, the structural stability of bone tissue directly affects the fixation effect and stability of the screw. The propagation distance index reflects whether the bone voxel is within a safe anatomical range and can be used for spatial constraints. Therefore, in this module, the bone connectivity and propagation distance index of each bone voxel can be fused to calculate the screw placement suitability score for each bone voxel, reflecting whether each bone voxel is suitable for screw placement.

[0053] Preferably, in one embodiment of the present invention, the method for obtaining the nail placement suitability score includes: As can be seen in the geometric field analysis module 102, the smaller the propagation distance index, the closer the bone tissue voxel is to the seed bone region (vertebral body core); conversely, the larger the value, the farther the bone tissue voxel is from the seed bone region (vertebral body core). Therefore, when the propagation distance index of a certain bone tissue voxel is less than the preset distance threshold, the distance cutoff value of the bone tissue voxel is set to 1, indicating that it is within the anatomical safety range and is considered to be within the area where surgery can be performed; otherwise, it is set to 0.

[0054] As shown in the connectivity analysis module 103, the greater the bone connectivity, the better the connectivity with the seed bone region. Since the bone connectivity is a value less than 1, exponential operation on it will produce a Matthew effect, where larger values ​​will remain large, while smaller values ​​will drop sharply. This can significantly widen the score gap between effective bone and free bone. Therefore, exponential operation is performed on the bone connectivity value of each bone tissue voxel to obtain the connectivity enhancement factor. At this time, the exponent is set to 2.

[0055] Finally, the connectivity enhancement factor of each bone tissue voxel is multiplied by the distance cutoff value to obtain the fixation suitability score for each bone tissue voxel. Based on the above analysis, it can be seen that the larger the fixation suitability score, the more likely the bone tissue voxel is within the anatomically safe range and has good connectivity, making it suitable as a fixation area.

[0056] It should be noted that the preset distance threshold is set to 60mm, but the specific value can be adjusted according to the implementation scenario and is not limited here.

[0057] After obtaining the fixation suitability score for each bone tissue voxel, effective channel voxels can be selected based on the score as suitable fixation locations.

[0058] Preferably, in one embodiment of the present invention, the method for obtaining effective channel voxels includes: To improve computational efficiency and eliminate low-value noise, bone tissue voxels with higher suitability need to be selected: All bone tissue voxels are sorted in ascending order of pin placement suitability scores, and the pin placement suitability score at a preset percentile threshold is used as the score selection threshold. Finally, bone tissue voxels with pin placement suitability scores greater than or equal to the preset score selection threshold are selected as valid channel voxels.

[0059] It should be noted that in this embodiment of the present invention, the preset percentile threshold is set to 85.

[0060] After obtaining the effective channel voxels, the pin placement path can be determined based on the location distribution characteristics of the effective channel voxels and the pin placement suitability score.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining the pin placement path includes: To plan the pedicle screw paths for the left and right sides respectively, the global point set needs to be anatomically segmented. Therefore, the geometric center of the seed bone region is obtained and a sagittal segmentation plane is constructed. This sagittal segmentation plane passes through the geometric center and its normal vector is parallel to the horizontal direction (X-axis direction). At this time, the effective channel voxels can be divided into two subsets, left and right, using the sagittal segmentation plane.

[0062] Then, in each subset, the path axis needs to be extracted: the coordinates of the effective channel voxels in the 3D CT image are weighted and averaged using the pin placement suitability score of the effective channel voxels to obtain the weighting center of each subset. ,in, The coordinates of the weighting center are indicated. This indicates the number of valid channel voxels in each subset; This represents the pinning suitability score of the j-th effective channel voxel in each subset; This represents the coordinates of the j-th valid channel voxel in each subset.

[0063] Since bone tissue voxels with high fixation suitability scores tend to cluster in areas of intact bone, the calculated weighted center will automatically shift towards areas of dense bone and good connectivity, thus avoiding the fracture zone.

[0064] Next, a covariance matrix is ​​constructed based on the three coordinate axes (X-axis, Y-axis, and Z-axis) in three-dimensional space. At each element position, a weighted average is calculated using the weighted center of each subset, the pinning suitability score of the effective channel voxels in each subset, and the coordinates of the effective channel voxels to obtain the value of each element in the covariance matrix. ;in, , representing the coordinate axes Let represent the element in the m-th row and n-th column of the covariance matrix. This represents the coordinates of the j-th valid channel voxel in each subset on the coordinate axis corresponding to m. This represents the coordinates of the weighted center of each subset on the coordinate axis corresponding to m. This represents the coordinates of the j-th effective channel voxel in each subset on the coordinate axis corresponding to n. This represents the coordinates of the weighted center of each subset on the coordinate axis corresponding to n.

[0065] The covariance matrix can describe the distribution characteristics and correlation of effective channel voxels in three-dimensional space. Eigenvalue decomposition is performed on the covariance matrix corresponding to each subset. The eigenvector corresponding to the largest eigenvalue represents the main direction of distribution of effective channel voxels. Therefore, the eigenvector corresponding to the largest eigenvalue is taken as the direction of entry into each subset.

[0066] Finally, a local cylindrical coordinate system was established based on the screw insertion direction, and the vertical distance distribution from the effective channel voxels to the straight line of the screw insertion direction was statistically analyzed to determine the maximum safe channel radius of the effective channel voxels in each subset: the vertical distances between all effective channel voxels in each subset and the straight line of the screw insertion direction were sorted in ascending order to obtain a distance sequence; in the distance sequence, the vertical distance at the preset percentile was taken as the maximum safe channel radius. The preset percentile was set to 90, which means that 90% of the highly suitable effective channel voxels are within this radius, ensuring that the screw will not penetrate the bone cortex.

[0067] At this point, a standard cylindrical model with a length spanning each subset and a radius equal to the maximum safe passage radius can be generated using visualization tools. The spinal skeleton model can then be displayed using volume rendering technology, and the cylindrical passage can be highlighted in green. This provides doctors with an intuitive virtual surgical passage that avoids the fracture line, assisting in preoperative planning and verification.

[0068] In summary, the data acquisition module obtains 3D CT images of the patient's spine, presenting the anatomical structure of the spine from all angles and perspectives. By filtering bone tissue voxels based on the numerical characteristics of their CT values, it can quickly separate bone tissue-related information from complex 3D image data. Then, in the geometric field analysis module, seed bone regions are determined based on the CT values ​​and location characteristics of the bone tissue voxels, enabling more accurate localization of representative and critical bone tissue areas and providing a reasonable starting point for subsequent anisotropic propagation analysis. Anisotropic propagation based on the CT values ​​of bone tissue voxels, obtaining the propagation distance index for each voxel, effectively prevents deviations from the correct anatomical range during subsequent path planning. Furthermore, to understand the microscopic continuity of bone structure, the connectivity analysis module analyzes the differences in CT values ​​between bone tissue voxels and the similarity of CT value changes to determine the connectivity index between any two bone tissue voxels, reflecting the continuity and integrity of the bone tissue. Based on the seed bone region, iterative calculations using connectivity indices between bone tissue voxels determine the bone connectivity of each voxel, avoiding free bone fragments and ensuring subsequent paths avoid unstable fracture areas. Finally, in the path planning module, based on the bone connectivity of each voxel and using propagation distance indices for spatial constraints, a screw placement suitability score is calculated for each voxel. Effective channel voxels are selected based on their placement suitability scores, and the screw placement path is determined according to the location distribution characteristics of these effective channel voxels and the placement suitability score. This fully utilizes the advantages of effective channel voxels and, combined with the patient's individualized spinal structure characteristics, plans the optimal screw placement path.

[0069] Please see Figure 2 This illustration shows a schematic diagram of the system structure of a three-dimensional visualization system for pedicle screw implantation path in spinal trauma according to an embodiment of the present invention. It includes a processor 200, a memory 201, a bus 202, and a communication interface 203. The processor 200, communication interface 203, and memory 201 are connected via the bus 202. The memory 201 may include a high-speed random access memory, and the bus 202 may be an ISA bus, PCI bus, or EISA bus, etc. The processor 200 may be an integrated circuit chip with signal processing capabilities. The memory 201 stores at least one instruction, at least one program, code set, or instruction set. When the processor loads and executes the at least one instruction, at least one program, code set, or instruction set, it implements the steps in the aforementioned modules.

[0070] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0072] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A three-dimensional visualization system for pedicle screw placement path in spinal trauma, characterized in that, The system includes: The data acquisition module is used to acquire three-dimensional CT images of the patient's spine; in the three-dimensional CT images, bone tissue voxels are selected based on the numerical characteristics of the CT values ​​of the voxel blocks. The geometric field analysis module is used to determine the seed bone region based on the CT value and location characteristics of bone tissue voxels and to perform anisotropic propagation based on the CT value of bone tissue voxels, thereby obtaining the propagation distance index of each bone tissue voxel. The connectivity analysis module is used to analyze the differences in CT values ​​and the similarity in changes of CT values ​​between bone tissue voxels to determine the connectivity index between bone tissue voxels. Based on the seed bone region, iterative calculations are performed based on the connectivity index between bone tissue voxels to determine the bone connectivity of each bone tissue voxel. The path planning module is used to calculate the pin placement suitability score for each bone tissue voxel based on the bone connectivity of each bone tissue voxel and to use the propagation distance index for spatial constraints, and to screen effective channel voxels; based on the location distribution characteristics of effective channel voxels and the pin placement suitability score, the pin placement path is determined.

2. The three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 1, characterized in that, The method for obtaining the seed bone region includes: Connectivity analysis was performed on all bone tissue voxels to obtain all connected components of the bone tissue. In each bone tissue connected region, the sum of the CT values ​​of all bone tissue voxels is calculated as the CT feature value of each bone tissue connected region; Among all bone tissue connected regions, the bone tissue connected region with the largest CT feature value is selected as the seed bone region.

3. The three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 1, characterized in that, The method for obtaining the propagation distance index includes: The propagation distance index of bone tissue voxels in the seed bone region is set to a first preset value, and the propagation distance index of bone tissue voxels in the non-seed bone region is set to a second preset value, wherein the first preset value is less than the second preset value. For each bone tissue voxel in the non-seed bone region, the sum of its CT value and a preset constant is used as the local propagation velocity. Starting from the bone tissue voxel in the seed bone region, the anisotropic propagation equation is solved using the fast travel algorithm to calculate the minimum cumulative propagation cost for the bone tissue voxel in the non-seed bone region to reach the bone tissue voxel in the seed bone region. This cost replaces the second preset value as a new propagation distance index for the bone tissue voxel in the non-seed bone region.

4. The three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 1, characterized in that, The method for obtaining the connectivity index includes: Determine the local neighborhood of each bone tissue voxel centered on each bone tissue voxel; Within the local neighborhood of each bone tissue voxel, the absolute value of the difference between the CT values ​​of the central voxel and each neighboring voxel is negatively correlated and normalized to obtain the similarity factor between the central voxel and each neighboring voxel. Calculate the CT value gradient modulus for each bone tissue voxel; Within the local neighborhood of each bone tissue voxel, the maximum value of the CT value gradient modulus between the central voxel and each neighboring voxel is negatively correlated and normalized, and this value is used as the connectivity factor between the central voxel and each neighboring voxel. The normalized value of the product of the similarity factor and the connectivity factor between the central voxel and each neighboring voxel is used as the connectivity index between each central voxel and each neighboring voxel.

5. A three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 4, characterized in that, The method for obtaining bone connectivity includes: The bone connectivity of each bone tissue voxel is initialized, and the initial values ​​of bone connectivity of bone tissue voxels in the seed bone region and the non-seed bone region are different. In each iteration, for any bone tissue voxel, the formulaic model for the bone connectivity of that voxel in that iteration includes: in, This indicates the bone connectivity of the bone tissue voxel in this iteration; Indicates the damping factor; This represents the initial value of bone connectivity for the voxel in the bone tissue. This indicates the number of neighboring voxels of the bone tissue voxel; This represents the bone connectivity of the i-th neighboring voxel of the bone tissue in the previous adjacent iteration; This represents the connectivity index between the bone tissue voxel and its i-th neighboring voxel; After each iteration, the bone connectivity of all bone tissue voxels is used to form a connectivity numerical vector. The iteration stops when the Euclidean distance between the connectivity numerical vectors of two adjacent iterations is less than a preset convergence threshold, and the bone connectivity of each bone tissue voxel in the last iteration is obtained.

6. A three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 1, characterized in that, The method for obtaining the staple placement suitability score includes: When the propagation distance index of a certain bone tissue voxel is less than the preset distance threshold, the distance cutoff value of the bone tissue voxel is set to 1; otherwise, it is set to 0. The connectivity enhancement factor is obtained by exponentially calculating the bone connectivity value of each bone tissue voxel. The connectivity enhancement factor of each bone tissue voxel is multiplied by the distance cutoff value to obtain the pin placement suitability score for each bone tissue voxel.

7. A three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 1, characterized in that, The method for obtaining the effective channel voxel includes: All bone tissue voxels were sorted in ascending order for nail placement suitability scores, and the nail placement suitability scores at the preset percentile threshold were used as the score screening threshold. Bone tissue voxels with a pin placement suitability score greater than or equal to a preset score screening threshold are considered as valid channel voxels.

8. A three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 1, characterized in that, The method for obtaining the pin placement path includes: Obtain the geometric center of the seed bone region and construct a sagittal segmentation plane, wherein the sagittal segmentation plane passes through the geometric center and its normal vector is parallel to the horizontal direction; The effective channel voxels are divided into two subsets using the sagittal segmentation plane; In each subset, the coordinates of the effective channel voxels in the 3D CT image are weighted and averaged using the pin placement suitability score of the effective channel voxels to obtain the weighting center of each subset. A covariance matrix is ​​constructed based on the three coordinate axes in three-dimensional space. At each element position, a weighted average is performed using the weighting center of each subset, the pinning suitability score of the effective channel voxel in each subset, and the coordinates of the effective channel voxel to obtain the value of each element in the covariance matrix. The covariance matrix corresponding to each subset is decomposed into eigenvalues, and the eigenvector corresponding to the largest eigenvalue is used as the nailing direction in each subset. A local cylindrical coordinate system is established based on the nail-feeding direction. The vertical distance distribution from the effective channel voxel set to the straight line containing the nail-feeding direction is statistically analyzed, and the maximum safe channel radius of the effective channel voxels in each subset is determined.

9. A three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 8, characterized in that, The method for obtaining the maximum safe passage radius includes: Arrange the vertical distances between all valid channel voxels in each subset and the straight line containing the nail insertion direction in ascending order to obtain a distance sequence; In the distance sequence, the vertical distance at the preset percentile is used as the maximum safe passage radius.

10. A three-dimensional visualization system for pedicle screw placement path in spinal trauma according to claim 1, characterized in that, The method for obtaining the bone tissue voxels includes: In the three-dimensional CT image, voxel blocks with CT values ​​greater than a preset CT threshold are taken as bone tissue voxels, wherein the CT value in the three-dimensional CT image is a normalized value, and the preset CT threshold ranges from [0.3, 0.5].

Images