A navigation milling method and surgical robot system for craniofacial abnormal proliferative bone fibers

Abstract

The application provides a navigation grinding method and a surgical robot system for abnormal proliferative bone fibers in the craniofacial region, S1: acquiring craniofacial bone tissue image data and intraoral scanning data of a target human body, and constructing a three-dimensional digital craniofacial model; S2: based on the abnormal pathological characteristics of the abnormal proliferative bone fibers in the craniofacial bone tissue image data, identifying and acquiring a lesion area of the abnormal proliferative bone fibers, and generating a three-dimensional model of the lesion area; S3: segmenting and acquiring key human tissue structures present in the craniofacial region, and calculating the minimum distance between the abnormal proliferative bone fibers in the lesion area of the abnormal proliferative bone fibers and any key human tissue structure, and generating a distance field map; S4: based on the three-dimensional model of the lesion area and the distance field map, establishing a navigation grinding path of a surgical robot, thereby realizing functions such as medical image data recognition and analysis, automatic generation of a navigation grinding path, and robot-assisted surgery based on artificial intelligence technology.

Description

Technical Field

[0001] This invention belongs to the field of medical device technology, and in particular relates to a navigation-guided grinding method and surgical robot system for abnormally proliferating bone fibers in the craniofacial region. Background Technology

[0002] Craniofacial fibrous dysplasia is a fibrous-osseous developmental disorder that occurs in the skull and maxillofacial bones. Pathologically, it is characterized by the replacement of normal bone tissue by cellular fibrous connective tissue and irregularly shaped trabecular bone. Clinically, it manifests as progressive, painless bone swelling, which can lead to facial asymmetry, malocclusion, visual impairment, hearing loss, and airway stenosis, among other functional impairments.

[0003] Existing surgical robot-guided abrasion methods for abnormally proliferating bone fibers in the craniofacial region often rely on manual control by medical personnel based on their own experience for abrasion path planning. This makes it extremely difficult to accurately determine the abrasion range, especially when it comes to determining the lesion boundary. Furthermore, in areas where the bone is deformed, the location of key human tissue structures is atypical, which can easily cause iatrogenic damage.

[0004] Therefore, under the current technology, there is a lack of a method and surgical robot system for digital acquisition, identification, navigation and grinding path planning of abnormal proliferating bone fibers in the craniofacial region that can be applied to surgical robots. Summary of the Invention

[0005] This invention provides a navigation-guided abrasion method and surgical robot system for abnormally proliferating bone fibers in the craniofacial region, in order to solve the technical problem that it is difficult to achieve accurate identification of abnormally proliferating bone fibers in the craniofacial region and the navigation-guided abrasion path planning function of the surgical robot through digital means under the existing technology.

[0006] To solve the above problems, the technical solution of the present invention is: a navigation-guided abrasion method for abnormally proliferating bone fibers in the craniofacial region, comprising the following steps: S1: Acquire craniofacial bone tissue image data and intraoral scan data of the target human body, and perform registration and fusion processing on the craniofacial bone tissue image data and the intraoral scan data in the same coordinate system to construct a three-dimensional digital craniofacial model containing craniofacial skeletal structure and dental arch structure. S2: In the three-dimensional digital craniofacial model, based on the abnormal pathological features of abnormally proliferating bone fibers in the craniofacial bone tissue image data, the lesion area of ​​fibrous dysplasia is identified and obtained, and a three-dimensional model of the lesion area is updated and generated based on the model of the three-dimensional digital craniofacial model. S3: In the three-dimensional model of the lesion area, segment and obtain the key human tissue structures existing in the craniofacial region, and calculate the minimum distance between the abnormally proliferating bone fibers in the fibrous dysplasia lesion area and any of the key human tissue structures to generate a distance field map. S4: Based on the three-dimensional model of the lesion area and the distance field map, establish the navigation and grinding path of the surgical robot.

[0007] Preferably, in S1, the craniofacial bone tissue image data and the intraoral scan data are registered and fused in the same standard coordinate system, including the following steps: S11: Select the bilateral maxillary tuberosities, zygomatic alveolar ridges, and cervical margins of teeth from the craniofacial bone tissue image data and the intraoral scan data as registration and fusion feature points to achieve alignment and registration of the craniofacial bone tissue image data and the intraoral scan data in the same coordinate system.

[0008] Preferably, in S2, based on the abnormal pathological features of the abnormally proliferating bone fibers in the craniofacial bone tissue imaging data, the lesion area of ​​the fibrous dysplasia is identified and obtained, including the following steps: S21: Calculate the normalized grayscale attenuation coefficient G for any voxel in the skeletal region of the craniofacial bone tissue image data. n The normalized grayscale value attenuation coefficient G n The calculation expression is: in, The CT value of the target voxel. and These represent the maximum and minimum CT values ​​within the skeletal region. The normalized gray value attenuation coefficient G n It is configured to indirectly distinguish between abnormally proliferating bone fibers and normal bone fibers in a skeletal region based on the effect of bone density on CT grayscale values. S22: Calculate the local texture entropy E for any voxel in the skeletal region of the craniofacial bone tissue image data. The expression for calculating the local texture entropy E is as follows: Among them, the target voxel is set Within the 3×3×3 neighborhood centered on, grayscale The probability of occurrence Limited to 256; The local texture entropy E is configured to determine the presence of abnormally proliferating bone fibers and normal bone fibers in the skeletal region based on the local texture complexity of abnormally proliferating bone fibers. S23: Calculate the local grayscale standard deviation for any voxel in the skeletal region of the craniofacial bone tissue image data. The local grayscale standard deviation The calculation expression is: Where N is the set of voxels in the neighborhood, The grayscale value fluctuates within the domain; The local gray standard deviation It is configured to indirectly distinguish between abnormally proliferating bone fibers and normal bone fibers in the skeletal region based on the effect of the uneven internal composition of abnormally proliferating bone fibers on the internal gray-scale fluctuation. S24: Calculate the local gradient magnitude Grad for any voxel in the skeletal region of the craniofacial bone tissue image data. The expression for calculating the local gradient magnitude Grad is as follows: The local gradient amplitude Grad is configured to distinguish between abnormally proliferating bone fibers and normal bone fibers in a skeletal region based on the lesion boundary of abnormally proliferating bone fibers.

[0009] Preferably, in S2, based on the abnormal pathological features of the abnormally proliferating bone fibers in the craniofacial bone tissue imaging data, the identification and acquisition of the lesion area of ​​fibrous dysplasia further includes the following steps: S25: Based on the normalized grayscale value attenuation coefficient G n The local texture entropy E, the local grayscale standard deviation Using the local gradient magnitude Grad, the fusion recognition probability of abnormally proliferating bone fibers is calculated. The expression for calculating the fusion recognition probability of abnormally proliferating bone fibers is as follows: in, , , These are the local texture entropy E and the local grayscale standard deviation, respectively. The local gradient magnitude Grad value after normalization to [0,1] , , , These are the weighting coefficients, and .

[0010] Preferably, the key human tissue structures include the infraorbital neurovascular bundle, the inferior alveolar nerve canal, and the optic nerve canal.

[0011] Preferably, in S4, based on the three-dimensional model of the lesion area and the distance field map, a navigation and abrasion path for the surgical robot is established, including the following steps: S41: Based on the three-dimensional model of the lesion area, calculate the abrasion depth L(p) of each voxel in the fibrous dysplasia lesion area. The calculation expression for the abrasion depth L(p) is as follows: in, The maximum safe depth of abrasion is determined based on the preoperative measurement of the cortical bone thickness in the lesion area. The normalized abnormality probability score is determined based on the fusion identification probability of abnormally proliferating bone fibers. This refers to the preset attenuation function for the protection of key human tissues and structures. The mathematical expression is: in, For minimum safe distance, This is the maximum safe distance.

[0012] Preferably, in S4, based on the three-dimensional model of the lesion area and the distance field map, a navigation and abrasion path for the surgical robot is established, further including the following steps: S42: Preset depth increment The baseline depth value of the bone surface in the fibrous dysplasia lesion region was set to 0, and different grinding depth values ​​in the fibrous dysplasia lesion region were extracted layer by layer. And generate contour lines of the grinding depth boundary; S43: Generate a surgical robot navigation grinding path along the contour lines of the grinding depth boundaries of several layers and using a smoothing algorithm to generate a continuous spiral distribution.

[0013] Preferably, in step S4, based on the three-dimensional model of the lesion area and the distance field map, the navigation and abrasion path of the surgical robot is established, which further includes the following steps: S44: Obtain the normal vector of a unit area of ​​the bone surface in the fibrous dysplasia lesion region. The normal vector of the unit area of ​​the bone surface is used to define the feed direction of the surgical robot end effector located at the reference depth value of the unit area. Between the reference depth value path points of the surgical robot navigation grinding path, the posture of the surgical robot end effector is smoothed and adjusted.

[0014] Preferably, the navigation-guided abrasion method for abnormally proliferating bone fibers in the craniofacial region further includes the following steps: S5: Based on bone density and grinding process, a contact force range is preset for the end effector of the surgical robot. During the movement of the surgical robot along the grinding path, if the actual contact force between the surgical robot end effector and the human skeleton is greater than the maximum contact force value, the feed speed of the surgical robot end effector is limited to slow down and the grinding depth is reduced; if the actual contact force between the surgical robot end effector and the human skeleton is less than the minimum contact force value, the feed speed of the surgical robot end effector is limited to increase.

[0015] Based on the same concept, the present invention also provides a navigation-guided ablation surgical robot system for craniofacial abnormal proliferative bone fibers, for performing the navigation-guided ablation method for craniofacial abnormal proliferative bone fibers as described above, comprising: The craniofacial morphology modeling module is used to acquire craniofacial bone tissue image data and intraoral scan data of the target human body, and to perform registration and fusion processing on the craniofacial bone tissue image data and the intraoral scan data in the same coordinate system to construct a three-dimensional digital craniofacial model containing craniofacial skeletal structure and dental arch structure. The lesion morphology modeling module is used to identify and obtain the lesion area of ​​fibrous dysplasia based on the abnormal pathological features of abnormally proliferating bone fibers in the craniofacial bone tissue image data in the three-dimensional digital craniofacial model, and update and generate a three-dimensional model of the lesion area based on the model of the three-dimensional digital craniofacial model. The safety constraint distance detection module is used to segment and obtain the key human tissue structures in the craniofacial region in the three-dimensional model of the lesion area, and calculate the minimum distance between the abnormally proliferating bone fibers in the fibrous dysplasia lesion area and any of the key human tissue structures, and generate a distance field map. The navigation and grinding path generation module is used to establish the navigation and grinding path of the surgical robot based on the three-dimensional model of the lesion area and the distance field map.

[0016] Because the present invention adopts the above technical solution, it has the following advantages and positive effects compared with the prior art: This invention provides a navigation-guided abrasion method and surgical robot system for abnormally proliferating bone fibers in the craniofacial region. First, it acquires craniofacial bone tissue imaging data and intraoral scan data of the target human body. The craniofacial bone tissue imaging data and intraoral scan data are registered and fused in the same coordinate system to construct a three-dimensional digital craniofacial model including the craniofacial skeletal structure and dental arch. In this three-dimensional digital craniofacial model, based on the abnormal pathological features of the abnormally proliferating bone fibers in the craniofacial bone tissue imaging data, the lesion area of ​​fibrous dysplasia is identified and obtained. Then, a three-dimensional model of the lesion area is updated and generated based on the three-dimensional digital craniofacial model. Subsequently, in the three-dimensional model of the lesion area, key human tissue structures present in the craniofacial region are segmented and obtained, and the minimum distance between the abnormally proliferating bone fibers in the fibrous dysplasia lesion area and any key human tissue structure is calculated to generate a distance field map. Finally, based on the three-dimensional model of the lesion area and the distance field map, a navigation abrasion path for the surgical robot is established. In this invention, the abnormal pathological characteristics of abnormally proliferating bone fibers are utilized to automatically determine the distribution location of these fibers through multi-feature fusion, achieving precise and repeatable detection. Simultaneously, based on a three-dimensional model of the lesion area, a navigation and abrasion path for the surgical robot is automatically generated. Furthermore, this process fully adapts to key human tissue structures, effectively improving the safety of the surgical robot in performing abrasion of abnormally proliferating bone fibers. Thus, the invention realizes functions such as medical image data recognition and analysis based on artificial intelligence technology, automatic generation of navigation and abrasion paths, and robot-assisted surgery. Attached Figure Description

[0017] Figure 1 This invention provides a schematic flowchart of a navigation-guided grinding method for abnormally proliferating bone fibers in the craniofacial region. Detailed Implementation

[0018] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a navigation-guided ablation method and surgical robot system for abnormally proliferating bone fibers in the craniofacial region, as proposed in this invention. The advantages and features of this invention will become more apparent from the following description and claims.

[0019] First Embodiment See Figure 1 This embodiment provides a navigation-guided abrasion method for abnormally proliferating bone fibers in the craniofacial region. Based on artificial intelligence and robot-assisted technology, it automatically detects the location of abnormally proliferating bone fibers through digital means, and generates a navigation-guided abrasion path to guide the movement of the surgical robot based on the location of the abnormally proliferating bone fibers and key human tissue structures.

[0020] First, it should be noted that the surgical robot described in this embodiment refers to a robotic device specifically designed for craniofacial fibrous dysplasia. It is equipped with at least one set of actuators that can move automatically along a preset path. The actuators are equipped with sawing, drilling, and grinding devices, such as reciprocating saws, oscillating saws, drill bits, and grinding drills, to reduce the cutting or grinding of bone tissue. This embodiment does not impose specific limitations on the device structure of the surgical robot.

[0021] The navigation-guided abrasion method for abnormally proliferating bone fibers in the craniofacial region specifically includes the following steps: S1: Obtain craniofacial bone tissue image data of the target human body based on CBCT or CT scanning technology, and obtain intraoral scan data of the target human body through intraoral scanning technology. Use data processing software (such as Mimics) to register and fuse the craniofacial bone tissue image data and intraoral scan data in the same coordinate system to construct a three-dimensional digital craniofacial model containing the craniofacial skeletal structure and dental arch structure. That is, the three-dimensional digital craniofacial model can be used to characterize the entire craniofacial skeletal morphology and dental occlusion relationship of the target human body.

[0022] S2: In the three-dimensional digital craniofacial model, abnormal proliferating bone fibers exhibit abnormal pathological features such as "abnormal grayscale", "disordered texture" and "blurred structure" on CT images compared to normal bone tissue. Therefore, based on the abnormal pathological features of abnormal proliferating bone fibers in craniofacial bone tissue imaging data, the lesion area of ​​fibrous dysplasia can be identified and obtained. Then, based on the three-dimensional digital craniofacial model, a three-dimensional model of the lesion area can be updated and generated. That is, the three-dimensional model of the lesion area can be used to characterize the three-dimensional distribution range of abnormal proliferating bone fibers.

[0023] S3: Based on CBCT or CT scanning technology, further segment and obtain the key human tissue structures in the craniofacial region in the three-dimensional model of the lesion area, and calculate the minimum distance between the abnormal proliferating bone fibers in the lesion area of ​​fibrous dysplasia of bone and any key human tissue structure to generate a distance field map. That is, the distance field map can be used to characterize the distance parameters between the abnormal proliferating bone fibers and the nearest key human tissue structure.

[0024] S4: Based on the three-dimensional model and distance field map of the lesion area, a navigation and grinding path for the surgical robot is established. In this embodiment, the navigation and grinding path of the surgical robot will take into account the distribution area of ​​abnormally proliferating bone fibers and the distribution location of key human tissue structures to improve the safety of the surgical robot during the moving and grinding process.

[0025] The following will provide a more detailed explanation of the specific implementation steps and functions of the navigation-guided abrasion method for abnormally proliferating bone fibers in the craniofacial region provided in this embodiment: Preferably, in one embodiment, the registration and fusion processing of craniofacial bone tissue image data and intraoral scan data in the same standard coordinate system in S1 includes the following steps: S11: Select the bilateral maxillary tuberosities, zygomatic alveolar ridges, and cervical margins of teeth from the craniofacial bone tissue imaging data and intraoral scan data as registration and fusion feature points to achieve alignment and registration of the craniofacial bone tissue imaging data and intraoral scan data in the same coordinate system.

[0026] The bilateral maxillary tuberosities are anatomically located in the posterior part of the maxilla, as round bony prominences distal to the third molars (wisdom teeth). The maxillary tuberosities are an inherent anatomical structure of the maxilla, and their basic morphology remains stable even with the loss of adjacent teeth. In fibrous dysplasia of bone, they are usually not the initial or core affected area. The zygomatic alveolar ridge is the bony ridge formed at the junction of the zygomatic bone and the maxilla. Similarly, the zygomatic alveolar ridge is a key load-bearing structure connecting the zygomatic bone and the maxilla, with high bone density and a robust morphology. It is usually not the initial or core affected area. The cervical margin of a tooth is located at the junction of the gingiva and the crown, forming a continuous, complex three-dimensional curve. The cervical margin of a tooth contains rich geometric information and can be used to accurately align the craniomaxillary skeletal model with the dental arch model. Therefore, in this embodiment, the bilateral maxillary tuberosities, zygomatic alveolar ridges, and cervical margins of teeth are selected as registration and fusion feature points. This can minimize the negative impact of abnormally proliferating bone fibers on the registration and fusion of craniofacial bone tissue imaging data and intraoral scan data, improve the alignment and registration accuracy of craniofacial bone tissue imaging data and intraoral scan data, and provide accurate original model data for the subsequent establishment of three-dimensional models of lesion areas and surgical robot navigation abrasion paths.

[0027] Furthermore, in one embodiment, in S2, based on the abnormal pathological features of abnormally proliferating bone fibers in craniofacial bone tissue imaging data, the lesion area of ​​fibrous dysplasia is identified and acquired, including the following steps: S21: Calculate the normalized grayscale attenuation coefficient G for any voxel in the skeletal region of the craniofacial bone tissue imaging data. n Normalized grayscale value attenuation coefficient G n The calculation expression is: in, The CT value of the target voxel. and These represent the maximum and minimum CT values ​​within the skeletal region. Normalized gray value attenuation coefficient G nThe algorithm is configured to indirectly distinguish between abnormal and normal bone fibers in a skeletal region based on the influence of bone density on CT grayscale values. Specifically, in this embodiment, since the lesion area of ​​fibrous dysplasia is characterized by abnormal bone fibers replacing normal bone tissue, the bone density of abnormal bone fibers is significantly lower than that of normal bone tissue. Therefore, the CT value (grayscale) of abnormal bone fibers is significantly lower than that of normal cortical bone, resulting in a higher normalized grayscale attenuation coefficient G. n The larger the value, the lower the gray level of the bone at that point, and the more likely it is to be abnormally proliferating bone fibers.

[0028] S22: Simultaneously, calculate the local texture entropy E for any voxel in the skeletal region of the craniofacial bone tissue image data. The expression for calculating the local texture entropy E is: Among them, the target voxel is set Within the 3×3×3 neighborhood centered on, grayscale The probability of occurrence Limited to 256; The local texture entropy E is configured to distinguish between abnormal and normal bone fibers in a skeletal region based on the local texture complexity of abnormally proliferating bone fibers. In this embodiment, abnormally proliferating bone fibers usually exhibit a "ground glass" physiological structure. The trabecular structure of abnormally proliferating bone fibers is disordered and arranged without direction. Its local texture complexity is much higher than that of structurally ordered normal bone or uniform cysts. Therefore, the higher the local texture entropy E value, the more disordered the bone texture is, and the more likely it is to be abnormally proliferating bone fibers.

[0029] S23: Simultaneously, calculate the local grayscale standard deviation for any voxel in the skeletal region of the craniofacial bone tissue imaging data. Local grayscale standard deviation The calculation expression is: Where N is the set of voxels in the neighborhood, The grayscale value fluctuates within the domain; Local grayscale standard deviation Configured to indirectly distinguish between abnormal and normal bone fibers in a skeletal region based on the effect of uneven internal composition of abnormally proliferating bone fibers on internal grayscale fluctuations. Specifically, in this embodiment, abnormally proliferating bone fibers are prone to containing a sponge-like structure or microcystic changes, resulting in greater grayscale fluctuations than uniform normal cancellous bone or single cortical bone. Therefore, the local grayscale standard deviation is higher. The higher the value, the greater the fluctuation in grayscale within the bone, and the more likely it is to be abnormally proliferating bone fibers.

[0030] S24: Simultaneously, calculate the local gradient magnitude Grad for any voxel in the skeletal region of the craniofacial bone tissue imaging data. The expression for calculating the local gradient magnitude Grad is: The local gradient amplitude Grad is configured to distinguish between abnormally proliferating bone fibers and normal bone fibers in the skeletal region based on the lesion boundary of abnormally proliferating bone fibers. In this embodiment, the abnormally proliferating bone fibers and normal bone tissue grow infiltratively without a sharp boundary. Therefore, in the lesion boundary region, the local gradient amplitude Grad value will be lower than that at the junction of normal bone and soft tissue. Thus, the lower the local gradient amplitude Grad value, the more likely the skeletal region is to have abnormally proliferating bone fibers.

[0031] Furthermore, in one embodiment, in S2, based on the abnormal pathological features of the abnormally proliferating bone fibers in the craniofacial bone tissue imaging data, the identification and acquisition of the lesion area of ​​fibrous dysplasia of bone further includes the following steps: S25: Based on the normalized gray value attenuation coefficient G n Local texture entropy E, local grayscale standard deviation Using the local gradient magnitude Grad, the fusion recognition probability of abnormally proliferating bone fibers is calculated. The expression for calculating the fusion recognition probability of abnormally proliferating bone fibers is as follows: in, , , These are the local texture entropy E and the local grayscale standard deviation, respectively. The local gradient magnitude Grad value after normalization to [0,1]. , , , These are the weighting coefficients, and .

[0032] In this embodiment, abnormal proliferating fibrous bone exhibits complex multi-feature changes on imaging, including decreased density, structural disorder, internal heterogeneity, and blurred boundaries. This embodiment uses linear weighted fusion of four complementary features to comprehensively evaluate whether a skeletal region exhibits the complete imaging spectrum of fibrous dysplasia, greatly improving the accuracy and robustness of identifying abnormal proliferating fibrous bone.

[0033] Preferably, in one embodiment, the key human tissue structures include the infraorbital neurovascular bundle, the inferior alveolar nerve canal, and the optic nerve canal.

[0034] The infraorbital neurovascular bundle is located within the maxilla, the inferior alveolar canal is located within the mandible, and the optic canal is located within the zygomatic bone and the supraorbital wall. When abnormally proliferating bone fibers are shaved down in the maxilla, mandible, or zygomatic bone, the adjacent infraorbital neurovascular bundle, inferior alveolar canal, and optic canal are easily damaged.

[0035] Therefore, in one embodiment, in S4, the navigation and grinding path of the surgical robot is established based on the three-dimensional model of the lesion area and the distance field map, including the following steps: S41: Based on the three-dimensional model of the lesion area, calculate the abrasion depth L(p) of each voxel in the fibrous dysplasia lesion area along the direction perpendicular to the craniomaxillary plane. The expression for calculating the abrasion depth L(p) is as follows: in, The maximum safe depth of abrasion is determined based on the preoperative measurement of the cortical bone thickness in the lesion area. The normalized abnormality probability score is determined based on the fusion identification probability of abnormally proliferating bone fibers. This refers to the preset attenuation function for the protection of key human tissues and structures. The mathematical expression is: in, For minimum safe distance, The maximum safe distance is when the end effector of the surgical robot is less than the minimum safe distance from critical human tissue structures. When the distance between the surgical robot's end effector and critical human tissue structures exceeds the maximum safe distance, the surgical robot will stop its grinding action. When the surgical robot is in a certain position, it will maintain the preset feed grinding action; when the end effector of the surgical robot is greater than the minimum safe distance from the critical tissue structure of the human body... And less than the maximum safe distance At that time, the surgical robot dynamically executes the feed grinding action according to a certain grinding depth attenuation ratio.

[0036] Furthermore, in one embodiment, the navigation and abrasion path of the surgical robot is established in S4 based on the three-dimensional model of the lesion area and the distance field map, further including the following steps: S42: Preset depth increment The baseline depth of the bone surface in the fibrous dysplasia lesion area was set to 0, and different grinding depth values ​​were extracted layer by layer in the fibrous dysplasia lesion area. It generates grinding depth boundary contour lines, and each grinding depth boundary contour line defines the boundary of the area to be ground at that specific depth.

[0037] S43: A continuous spiral distribution surgical robot navigation grinding path is generated along several layers of grinding depth boundary contour lines and through a smoothing algorithm. The surgical robot can perform continuous and stable grinding along the navigation grinding path, reducing start-stop vibration and improving grinding efficiency.

[0038] Meanwhile, in one embodiment, the navigation and abrasion path of the surgical robot is established in S4 based on the three-dimensional model of the lesion area and the distance field map, which also includes the following steps: S44: Obtain the normal vector of the unit area of ​​the bone surface in the fibrous dysplasia lesion region (i.e., the area covered by a single grinding operation of the surgical robot end effector). The normal vector of the unit area of ​​the bone surface is used to define the feed direction of the surgical robot end effector located at the reference depth value of the unit area. The axis of the surgical robot end effector is aligned with this normal vector. Therefore, between the reference depth value path points of the surgical robot navigation grinding path, the posture of the surgical robot end effector is smoothed and adjusted to ensure that the posture change of the surgical robot end effector is continuous and stable during the movement, avoids violent shaking, and always maintains a perpendicular relationship with the bone surface.

[0039] Preferably, in one embodiment, the navigation-guided abrasion method for abnormally proliferating bone fibers in the craniofacial region further includes the following steps: S5: Based on bone density and grinding process, a contact force range is preset for the end effector of the surgical robot. During the movement of the surgical robot along the grinding path, if the actual contact force between the surgical robot's end effector and the human bone is greater than the maximum contact force value, it indicates that the grinding resistance is too high. It may encounter dense bone tissue that was not identified during the planning or instrument dulling. In this case, the feed speed of the surgical robot's end effector can be slowed down and the grinding depth reduced. If the actual contact force between the surgical robot's end effector and the human bone is less than the minimum contact force value, it means that the bone in that area is softer than expected (such as the core of a lesion). In this case, the original path can be maintained and the feed speed of the surgical robot's end effector can be appropriately increased.

[0040] In this embodiment, the bone heterogeneity of abnormally proliferating bone fibers is a typical feature. The force feedback control method provided in this embodiment provides "tactile" perception for the movement of the surgical robot, enabling it to adapt to this microscopic uncertainty, making up for the shortcomings of pure vision planning, and greatly improving the safety of using the surgical robot.

[0041] Second Embodiment Based on the same concept, this embodiment also provides a navigation-guided abrasion surgical robot system for abnormally proliferating bone fibers in the craniofacial region, used to perform the navigation-guided abrasion method for abnormally proliferating bone fibers in the craniofacial region as described in any one of the first embodiments, including: The craniofacial morphology modeling module is used to acquire craniofacial bone tissue image data and intraoral scan data of the target human body. The craniofacial bone tissue image data and intraoral scan data are registered and fused in the same coordinate system to construct a three-dimensional digital craniofacial model that includes the craniofacial skeletal structure and dental arch structure.

[0042] The lesion morphology modeling module is used to identify and obtain the lesion area of ​​fibrous dysplasia of bone based on the abnormal pathological features of abnormally proliferating bone fibers in craniofacial bone tissue imaging data in a three-dimensional digital craniofacial model, and update and generate a three-dimensional model of the lesion area based on the three-dimensional digital craniofacial model.

[0043] The safety constraint distance detection module is used to segment and obtain key human tissue structures in the craniofacial region in the 3D model of the lesion area, and calculate the minimum distance between the abnormally proliferating bone fibers in the fibrous dysplasia lesion area and any key human tissue structure, generating a distance field map.

[0044] The navigation and grinding path generation module is used to establish the navigation and grinding path of the surgical robot based on the 3D model of the lesion area and the distance field map.

[0045] The functions of each component in the above-mentioned navigation-guided abrasion surgical robot system for abnormally proliferating bone fibers in the craniofacial region correspond to the steps in the above-mentioned navigation-guided abrasion method embodiment for abnormally proliferating bone fibers in the craniofacial region. Their functions and implementation processes will not be described in detail here.

[0046] This embodiment also provides an electronic device, including a processor and a memory. The memory stores machine-executable instructions that can be executed by the processor. The processor executes the machine-executable instructions to implement the above-described navigation grinding method for abnormally proliferating bone fibers in the craniofacial region.

[0047] This embodiment also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the steps of the above-described navigation grinding method for abnormally proliferating bone fibers in the craniofacial region.

[0048] In summary, this invention provides a navigation-guided abrasion method and surgical robot system for abnormally proliferating bone fibers in the craniofacial region. First, it acquires craniofacial bone tissue imaging data and intraoral scan data of the target human body. Then, it performs registration and fusion processing on the craniofacial bone tissue imaging data and intraoral scan data in the same coordinate system to construct a three-dimensional digital craniofacial model including the craniofacial skeletal structure and dental arch structure. In this three-dimensional digital craniofacial model, based on the abnormal pathological features of the abnormally proliferating bone fibers in the craniofacial bone tissue imaging data, it identifies and obtains the lesion area of ​​fibrous dysplasia. Then, it updates and generates a three-dimensional model of the lesion area based on the three-dimensional digital craniofacial model. Subsequently, in the three-dimensional model of the lesion area, it segments and obtains key human tissue structures present in the craniofacial region, and calculates the minimum distance between the abnormally proliferating bone fibers in the fibrous dysplasia lesion area and any key human tissue structure, generating a distance field map. Finally, based on the three-dimensional model of the lesion area and the distance field map, it establishes the navigation abrasion path for the surgical robot. In this invention, the abnormal pathological characteristics of abnormally proliferating bone fibers are utilized to automatically determine their distribution location through multi-feature fusion. After accurately identifying the boundaries of the abnormally proliferating bone fibers, the midline of the face is determined using artificial intelligence technology. Subsequently, the bone contour shape that the surgical robot needs to modify is generated and confirmed by referring to the surface morphology of the healthy side, achieving precise and repeatable detection functions. At the same time, based on the three-dimensional model of the lesion area, a navigation and grinding path for the surgical robot is automatically generated. In this process, the system is fully adapted to key human tissue structures, which can effectively improve the safety of the surgical robot in performing abnormally proliferating bone fiber grinding operations. Thus, the invention realizes functions such as medical image data recognition and analysis, automatic generation of navigation and grinding paths, and robot-assisted surgery based on artificial intelligence technology.

[0049] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they shall still fall within the protection scope of the present invention.

Claims

1. A navigation-guided ablation surgical robot system for abnormally proliferating bone fibers in the craniofacial region, characterized in that, include: The craniofacial morphology modeling module is used to acquire craniofacial bone tissue image data and intraoral scan data of the target human body, and to perform registration and fusion processing on the craniofacial bone tissue image data and the intraoral scan data in the same coordinate system to construct a three-dimensional digital craniofacial model containing craniofacial skeletal structure and dental arch structure. The lesion morphology modeling module is used to identify and obtain the lesion area of ​​fibrous dysplasia based on the abnormal pathological features of abnormally proliferating bone fibers in the craniofacial bone tissue image data in the three-dimensional digital craniofacial model, and update and generate a three-dimensional model of the lesion area based on the model of the three-dimensional digital craniofacial model. The safety constraint distance detection module is used to segment and obtain the key human tissue structures in the craniofacial region in the three-dimensional model of the lesion area, and calculate the minimum distance between the abnormally proliferating bone fibers in the fibrous dysplasia lesion area and any of the key human tissue structures, and generate a distance field map. A navigation abrasion path generation module is used to establish a navigation abrasion path for the surgical robot based on the 3D model of the lesion area and the distance field map; wherein, the navigation abrasion path generation module is configured as follows: S41: Based on the three-dimensional model of the lesion area, calculate the abrasion depth L(p) of each voxel in the fibrous dysplasia lesion area. The calculation expression for the abrasion depth L(p) is as follows: in, The maximum safe depth of abrasion is determined based on the preoperative measurement of the cortical bone thickness in the lesion area. The normalized abnormality probability score is determined based on the fusion identification probability of abnormally proliferating bone fibers. This refers to the preset attenuation function for the protection of key human tissues and structures. The mathematical expression is: in, For minimum safe distance, The maximum safe distance; S42: Preset depth increment The baseline depth value of the bone surface in the fibrous dysplasia lesion region was set to 0, and different grinding depth values ​​in the fibrous dysplasia lesion region were extracted layer by layer. And generate contour lines of the grinding depth boundary; S43: Generate a continuously spiral-distributed surgical robot navigation grinding path along the contour lines of the grinding depth boundaries of several layers and through a smoothing algorithm; S44: Obtain the normal vector of a unit area of ​​the bone surface in the fibrous dysplasia lesion region. The normal vector of the unit area of ​​the bone surface is used to define the feed direction of the surgical robot end effector located at the reference depth value of the unit area. Between the reference depth value path points of the surgical robot navigation grinding path, the posture of the surgical robot end effector is smoothed and adjusted.

2. The navigation-guided ablation surgical robot system for abnormally proliferating bone fibers in the craniofacial region as described in claim 1, characterized in that, The craniofacial bone tissue image data and the intraoral scan data were registered and fused in the same standard coordinate system, and configured as follows: S11: Select the bilateral maxillary tuberosities, zygomatic alveolar ridges, and cervical margins of teeth from the craniofacial bone tissue image data and the intraoral scan data as registration and fusion feature points to achieve alignment and registration of the craniofacial bone tissue image data and the intraoral scan data in the same coordinate system.

3. The navigation-guided ablation surgical robot system for abnormally proliferating bone fibers in the craniofacial region as described in claim 2, characterized in that, Based on the abnormal pathological features of abnormally proliferating bone fibers in the craniofacial bone tissue imaging data, the lesion area of ​​the bone fibrous dysplasia is identified and acquired, and configured as follows: S21: Calculate the normalized grayscale attenuation coefficient G for any voxel in the skeletal region of the craniofacial bone tissue image data. n The normalized grayscale value attenuation coefficient G n The calculation expression is: in, The CT value of the target voxel. and These represent the maximum and minimum CT values ​​within the skeletal region. The normalized gray value attenuation coefficient G n It is configured to indirectly distinguish between abnormally proliferating bone fibers and normal bone fibers in a skeletal region based on the effect of bone density on CT grayscale values. S22: Calculate the local texture entropy E for any voxel in the skeletal region of the craniofacial bone tissue image data. The expression for calculating the local texture entropy E is as follows: Among them, the target voxel is set Within the 3×3×3 neighborhood centered on, grayscale The probability of occurrence Limited to 256; The local texture entropy E is configured to determine the presence of abnormally proliferating bone fibers and normal bone fibers in the skeletal region based on the local texture complexity of abnormally proliferating bone fibers. S23: Calculate the local grayscale standard deviation for any voxel in the skeletal region of the craniofacial bone tissue image data. The local grayscale standard deviation The calculation expression is: Where N is the set of voxels in the neighborhood, The grayscale value fluctuates within the domain; The local gray standard deviation It is configured to indirectly distinguish between abnormally proliferating bone fibers and normal bone fibers in the skeletal region based on the effect of the uneven internal composition of abnormally proliferating bone fibers on the internal gray-scale fluctuation. S24: Calculate the local gradient magnitude Grad for any voxel in the skeletal region of the craniofacial bone tissue image data. The expression for calculating the local gradient magnitude Grad is as follows: The local gradient amplitude Grad is configured to distinguish between abnormally proliferating bone fibers and normal bone fibers in a skeletal region based on the lesion boundary of abnormally proliferating bone fibers.

4. The navigation-guided ablation surgical robot system for abnormally proliferating bone fibers in the craniofacial region as described in claim 3, characterized in that, Based on the abnormal pathological features of abnormally proliferating bone fibers in the craniofacial bone tissue imaging data, the lesion area of ​​the bone fibrous dysplasia is identified and acquired, and further configured as follows: S25: Based on the normalized grayscale value attenuation coefficient G n The local texture entropy E, the local grayscale standard deviation Using the local gradient magnitude Grad, the fusion recognition probability of abnormally proliferating bone fibers is calculated. The expression for calculating the fusion recognition probability of abnormally proliferating bone fibers is as follows: in, , , These are the local texture entropy E and the local grayscale standard deviation, respectively. The local gradient magnitude Grad value after normalization to [0,1] , , , These are the weighting coefficients, and .

5. The navigation-guided ablation surgical robot system for abnormally proliferating bone fibers in the craniofacial region as described in claim 1, characterized in that, The key human tissue structures mentioned include the infraorbital neurovascular bundle, the inferior alveolar canal, and the optic canal.

6. The navigation-guided ablation surgical robot system for abnormally proliferating bone fibers in the craniofacial region as described in claim 1, characterized in that, It is also configured as: Based on bone density and grinding process, a contact force range is preset for the end effector of the surgical robot. During the movement of the surgical robot along the grinding path, if the actual contact force between the surgical robot end effector and the human skeleton is greater than the maximum contact force value, the feed speed of the surgical robot end effector is limited to slow down and the grinding depth is reduced; if the actual contact force between the surgical robot end effector and the human skeleton is less than the minimum contact force value, the feed speed of the surgical robot end effector is limited to increase.

Images