Orthognathic surgery robot system

Abstract

The invention discloses a orthognathic surgery robot system, which comprises a multi-modal data fusion module, a cone beam CT (Computed Tomography) and intraoral scanning are integrated, a high-resolution three-dimensional bony structure image is acquired through the cone beam CT to reconstruct a three-dimensional anatomical model of maxillofacial skeletons, submillimeter details of tooth surfaces and occlusal relationships are captured through the intraoral scanning, and the submillimeter details of the tooth surfaces and occlusal relationships are acquired through the intraoral scanning; constructing a functional occlusion model; the navigation module is used for planning a navigation path through a preset self-adaptive path planning algorithm based on the three-dimensional anatomical model and the functional occlusion model of the maxillofacial skeleton of the target object; and the mechanical arm execution module is used for executing removal, osteotomy and fixation actions based on the navigation path. Through the multi-modal data fusion module, a personalized operation scheme can be formulated according to the specific anatomical structure and occlusal relationship of each patient.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical robots, and particularly relates to a orthognathic surgery robot system Background Art

[0002] ‌In the surgical method for bimaxillary protrusion, subapical osteotomy and setback of the upper and lower jaws after tooth extraction is an effective treatment method. This surgical method is mainly aimed at patients with bimaxillary protrusion, especially those adults with severe symptoms. During the operation, it is first necessary to extract the anterior teeth of the upper and lower jaws (usually the first bicuspid teeth) to create the necessary space for osteotomy and setback. Subsequently, subapical osteotomy is performed on the anterior parts of the upper and lower jaws, that is, the bone at the root of the teeth is transected. After the osteotomy is completed, the upper and lower jaw segments are adjusted backward to achieve the ideal dentomaxillary relationship. Finally, the adjusted bone segments are stabilized in the new position through a fixing device to ensure the surgical effect and promote bone healing. The upper anterior osteotomy usually requires extracting the first or second premolars on both sides of the upper jaw during the operation, then performing osteotomy within this gap, and finally retracting and fixing the cut bone. For the lower jaw, anterior subapical osteotomy of the mandible may be used, that is, anterior subapical osteotomy of the mandible is performed and the corresponding backward adjustment is carried out. The main risks of bone removal in bimaxillary protrusion surgery include bleeding, nerve injury, mandibular fracture, and uneven mandibular angle, etc.‌

[0003] Bimaxillary protrusion surgery is a relatively complex and risky surgical operation in oral and maxillofacial surgery. During the operation, osteotomy of the maxilla and mandible may be required, and this step may lead to bleeding, especially during osteotomy, more bleeding situations may be encountered. In addition, there is also a risk of nerve injury during the operation, which may cause numbness in the lower lip after surgery. Especially when removing excess bone mass during upper and lower jaw tooth extraction, the conventional surgical power system is difficult to accurately remove the excess bone mass, often increasing the surgical trauma. Summary of the Invention

[0004] To solve the above problems, the purpose of the present invention is to provide an orthognathic surgery robot system The technical solution provided by the present invention is: an orthognathic surgery robot system, comprising: A multimodal data fusion module, integrating cone beam CT and intraoral scanning, obtaining high-resolution three-dimensional bony structure images through the cone beam CT to reconstruct a three-dimensional anatomical model of the maxillofacial bones, and capturing submillimeter-level details of the tooth surface and occlusal relationship through the intraoral scanning to construct a functional occlusal model; A navigation module, planning a navigation path based on the three-dimensional anatomical model and functional occlusal model of the maxillofacial bones of the target object through a preset adaptive path planning algorithm; The robotic arm execution module performs removal, osteotomy, and fixation actions based on the navigation path.

[0005] Preferably, the cone beam CT scans the target object with cone beam X-rays to generate a three-dimensional model of hard tissues including at least the jawbone and temporomandibular joint, with an accuracy of ±0.2 mm, for preoperative planning of the osteotomy line position and the bone block movement path, and quantifies the local bone density through gray value analysis to provide biomechanical parameters for the power adjustment of the ultrasonic bone scalpel to avoid intraoperative bone fracture or insufficient cutting.

[0006] Preferably, the intraoral scan uses blue / white structured light scanning technology to obtain the surface morphology, adjacency relationship, and dental arch curve of the tooth crown, with a reconstruction accuracy of ±20 μm, for simulating the distribution of postoperative occlusal contact points on a virtual articulator.

[0007] Preferably, the preset adaptive path planning algorithm of the navigation module is the biomechanics-guided artificial potential field method, and the construction of the field function of the biomechanics-guided artificial potential field method includes: Attractive force field: Taking the preoperatively planned osteotomy line as the target point, the potential field strength is positively correlated with the bone density, and the formula is: U_att = 0.5k(ρ)⋅||q - q_goal||², k ∈ [50, 200] N / mm, q represents the real-time pose of the end effector of the robotic arm (or surgical tool), and q_goal represents the preoperatively planned osteotomy target pose; Repulsive force field: Blood vessels / nerves: Gaussian potential field, peak intensity 500 N / mm², action radius 3 mm; Soft tissue protection: A dynamic potential field that exponentially increases with the deformation amount; The path generation of the biomechanics-guided artificial potential field method includes: Potential field gradient descent method combined with virtual fixture: Set "non-traversable plane" constraints in sensitive areas, and enable the "tunnel guidance" mode for the thick bone area of the mandibular angle.

[0008] Preferably, the robotic arm execution module performing removal, osteotomy, and fixation actions based on the navigation path further includes: Minimally invasive extraction of the maxillary and mandibular first premolars with an ultrasonic bone scalpel to preserve the integrity of the alveolar septum bone wall; Piezoelectric bone scalpel performs osteotomy at a level 3 - 5 mm below the root apex to keep the maxillary part targeted at the nasal floor parallel line and the mandibular part targeted at the lower edge of the mental foramen, and the osteotomy depth is the unilateral cortical bone + part of the cancellous bone; Longitudinally osteotomize along the mesial and distal sides of the extraction socket to form a "bone block door" to protect the roots of adjacent teeth, with a safety distance ≥1.5 mm; Bluntly dissect the bone block from the labial mucoperiosteum, and retain the lingual soft tissue pedicle to preserve the blood supply source.

[0009] Preferably, a detachable orthognathic osteotomy milling cutter assembly is configured at the distal end of the robotic arm execution module. The orthognathic osteotomy milling cutter assembly includes a rod portion, a milling cutter, and an exclusion channel. The proximal end of the rod portion is detachably connected to the power output end of the orthognathic surgical robot system; the milling cutter includes a central axis support portion and a plurality of milling portions. The plurality of milling portions are circumferentially arranged on the central axis support portion and are evenly distributed; an exclusion channel is arranged between the plurality of milling portions to timely discharge the osteotomy bone fragments, and the maximum radius dimension of the milling cutter is coupled with the target osteotomy width.

[0010] Preferably, the milling portion includes a first extension portion and a second extension portion, and the second extension portion of one of the milling portions and the first extension portion of the adjacent milling portion form the exclusion channel.

[0011] Preferably, the navigation module performs real-time registration during the operation, combines the three-dimensional anatomical model of the maxillofacial bone and the intraoperative dynamic grating scanning data, and realizes the real-time spatial matching of the bony landmark points to correct the registration drift caused by the body position change.

[0012] Preferably, it further includes a bone healing evaluation model. The input data of the bone healing evaluation model includes image data, biological signal data, and mechanical signal data. The fusion module of the bone healing evaluation model adopts a cross-module attention mechanism to determine the parallel weights of the image data, biological signal data, and mechanical signal data, and generates a fusion feature vector; the time axis deviation of different modalities is aligned by dynamic time warping.

[0013] Due to the adoption of the above technical solutions, the present invention has the following advantages and positive effects compared with the prior art: The present invention utilizes cone-beam CT to obtain high-resolution three-dimensional bony structure images, enabling doctors to more accurately understand the craniofacial skeletal anatomical structure of patients. This helps in detailed pre-operative planning and simulation, reducing uncertainties during surgery. Intraoral scanning technology can capture sub-millimeter details of tooth surfaces and occlusion relationships, constructing a functional occlusion model. This helps ensure a more accurate occlusion relationship for patients after surgery, thereby increasing the success rate of the surgery and patient satisfaction. Through the multi-modal data fusion module, personalized surgical plans can be formulated according to the specific anatomical structure and occlusion relationship of each patient. This personalized treatment method can better meet the needs of different patients and improve the treatment effect. The navigation module, based on the stereoscopic anatomical model and functional occlusion model of the craniofacial bones of the target object, plans the navigation path through a preset adaptive path planning algorithm. This helps avoid possible accidents during the surgery and reduces the surgical risk. The robotic arm execution module can precisely perform actions such as removal, osteotomy, and fixation, reducing errors caused by manual operations and further reducing the surgical risk. By integrating advanced imaging technologies and intelligent algorithms, precise control of the surgical process and personalized treatment are achieved, significantly improving the precision, safety, and efficiency of the surgery, which is of great significance for promoting the development of medical technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The following further details the specific embodiments of the present invention in conjunction with the drawings, where: Figure 1 is a schematic diagram of the system composition of the orthognathic surgery robot system of the present invention; Figure 2 is a schematic diagram of the structure of the orthognathic osteotomy milling cutter assembly of the present invention.

[0015] Description of the reference numerals: 1 - milling cutter; 2 - rod part; 3 - discharge channel. SPECIFIC EMBODIMENTS

[0016] The following further details the present invention in conjunction with the drawings and specific embodiments. The advantages and features of the present invention will be clearer according to the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise ratios, only for conveniently and clearly assisting in explaining the purpose of the embodiments of the present invention.

[0017] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the drawings). If the specific posture changes, the directional indications will also change accordingly.

[0018] First Embodiment This embodiment provides an orthognathic surgery robot system, including: A multimodal data fusion module that integrates cone beam CT and intraoral scanning. The high-resolution three-dimensional bony structure image is obtained through the cone beam CT to reconstruct the three-dimensional anatomical model of the maxillofacial bones. The submillimeter-level details of the tooth surface and occlusion relationship are captured through the intraoral scanning to construct a functional occlusion model; A navigation module that plans a navigation path based on the three-dimensional anatomical model and the functional occlusion model of the target object's maxillofacial bones through a preset adaptive path planning algorithm; A robotic arm execution module that performs removal, osteotomy, and fixation actions based on the navigation path.

[0019] The technical solution of the present invention uses cone beam CT to obtain high-resolution three-dimensional bony structure images, enabling doctors to more accurately understand the maxillofacial bone anatomical structure of patients. This helps in detailed preoperative planning and simulation, reducing uncertainties during the operation. Intraoral scanning technology can capture submillimeter-level details of the tooth surface and occlusion relationship, constructing a functional occlusion model. This helps ensure a more accurate occlusion relationship for patients after the operation, thereby improving the success rate of the operation and patient satisfaction. Through the multimodal data fusion module, a personalized surgical plan can be formulated according to the specific anatomical structure and occlusion relationship of each patient. This personalized treatment method can better meet the needs of different patients and improve the treatment effect. The navigation module plans a navigation path based on the three-dimensional anatomical model and the functional occlusion model of the target object's maxillofacial bones through a preset adaptive path planning algorithm. This helps avoid unexpected situations that may occur during the operation and reduces the surgical risk. The robotic arm execution module can accurately perform actions such as removal, osteotomy, and fixation, reducing errors caused by manual operations and further reducing the surgical risk. By integrating advanced imaging technologies and intelligent algorithms, precise control and personalized treatment of the surgical process are achieved, significantly improving the accuracy, safety, and efficiency of the operation, which is of great significance for promoting the development of medical technologies.

[0020] In an embodiment of the present invention, the cone beam CT scans the target object with cone beam X-rays to generate a three-dimensional model of hard tissues including at least the jawbone and temporomandibular joint, with an accuracy of ±0.2 mm, for preoperative planning of the osteotomy line position and bone block movement path, and quantifying the local bone density through gray value analysis to provide biomechanical parameters for the power adjustment of the ultrasonic bone knife to avoid intraoperative bone fracture or insufficient cutting.

[0021] Cone-beam CT scans the target object using cone-beam X-rays and can generate a three-dimensional model with an accuracy of up to ±0.2 mm, covering important hard tissues such as the jawbone and temporomandibular joint. This high-precision three-dimensional model provides an extremely accurate basis for preoperative planning. Doctors can clearly understand the craniofacial bone structure of the patient, just like having an accurate "map". With this accurate model, doctors can determine the optimal osteotomy line position and bone block movement path in advance. In traditional surgeries, doctors often rely on two-dimensional images and experience for judgment, which has certain errors and uncertainties. In the present invention, through accurate three-dimensional model planning, the blindness during the surgery can be reduced, the accuracy of the surgery can be greatly improved, the surgery can be ensured to proceed as expected, and problems such as surgical failure or poor results caused by inaccurate planning can be effectively avoided. By analyzing the gray values of the three-dimensional model, the local bone density can be quantified. This function provides key biomechanical parameters for the power adjustment of the ultrasonic bone scalpel. Doctors can accurately adjust the power of the ultrasonic bone scalpel according to the bone density conditions of different parts. During the surgery, an appropriate power of the ultrasonic bone scalpel is crucial. If the power is too high, it is easy to cause bone fractures; if the power is too low, it may result in insufficient cutting. The biomechanical parameters provided by the gray value analysis in the present invention can provide a scientific basis for doctors, enabling them to flexibly adjust the power according to the actual situation, thereby effectively avoiding complications such as intraoperative bone fractures or insufficient cutting and ensuring the safety of the surgery.

[0022] In an embodiment of the present invention, the intraoral scan uses blue light / white light structured light scanning technology to obtain the surface morphology, adjacent relationship, and dental arch curve of the tooth crown, with a reconstruction accuracy of up to ±20 μm, for use in virtual articulator simulation of the postoperative occlusal contact point distribution.

[0023] Using blue light / white light structured light scanning technology, the reconstruction accuracy of the surface morphology, adjacent relationship, and dental arch curve of the tooth crown can reach ±20 μm. The high-precision data acquisition provides very accurate and detailed information for subsequent surgical planning and simulation, just like creating an extremely accurate "navigation map" for the entire surgical process. The accurate acquisition of the tooth crown surface morphology helps doctors deeply understand the tooth details of the patient before the surgery. The morphology of each tooth crown is accurately recorded, including its unique contour, protrusions, and depressions and other fine structures. These information can help doctors operate better during the surgery. In terms of the adjacent relationship, the accuracy of ±20 μm can clearly show the contact mode, angle, and tightness between teeth. This is very crucial for adjusting the tooth position and occlusal relationship during the surgery. Doctors can plan in advance how to make the teeth reach a more ideal adjacent state after the surgery based on these accurate data, avoiding oral problems caused by improper adjacency.

[0024] In an embodiment of the present invention, the preset adaptive path planning algorithm of the navigation module is a biomechanics-guided artificial potential field method. The construction of the field function of the biomechanics-guided artificial potential field method includes: Attractive force field: Taking the pre-operative planned osteotomy line as the target point, the potential field strength is positively correlated with bone density. The formula is: U_att = 0.5k(ρ)⋅||q - q_goal||², k ∈ [50, 200] N / mm, where q represents the real-time pose of the end effector of the robotic arm (or surgical tool), and q_goal represents the pre-operative planned osteotomy target pose; Repulsive force field: Blood vessels / nerves: Gaussian potential field, peak intensity 500 N / mm², action radius 3 mm; Soft tissue protection: A dynamic potential field that grows exponentially with the deformation amount; The path generation of the biomechanics-guided artificial potential field method includes: The potential field gradient descent method combined with virtual fixtures: Set "non-traversable plane" constraints in sensitive areas and enable the "tunnel guidance" mode for the thick bone area of the mandibular angle.

[0025] In the embodiment of the present invention, the attractive force field takes the pre-operative planned osteotomy line as the target point, and the potential field strength is positively correlated with bone density. This design enables the end effector of the robotic arm (or surgical tool) to adjust the movement path according to the bone density information of different parts in the complex maxillofacial bone environment. For example, in areas with higher bone density, the attractive force field strength is greater, prompting the actuator to move more accurately towards the target point. In the formula U_att = 0.5k(ρ)⋅||q - q_goal||², q represents the real-time pose of the end effector of the robotic arm (or surgical tool), and q_goal represents the pre-operative planned osteotomy target pose. The attractive force field calculated by this formula can guide the actuator to move from the current position to the preset osteotomy target pose. This calculation method based on real-time pose and target pose ensures that during the operation, even if the patient's head or bones move slightly, the actuator can always move towards the correct target, effectively improving the accuracy of the operation.

[0026] In an embodiment of the present invention, the robotic arm execution module further includes performing removal, osteotomy, and fixation actions based on the navigation path: Minimally invasive extraction of the maxillary and mandibular first premolars with an ultrasonic osteotome, preserving the integrity of the alveolar septum bone wall; Osteotomy at a level 3 - 5 mm below the root apex with a piezoelectric osteotome to keep the maxillary part targeted at the nasal floor parallel line and the mandibular part targeted at the lower edge of the mental foramen. The osteotomy depth is the unilateral cortical bone + part of the cancellous bone; Longitudinal osteotomy along the mesial and distal sides of the extraction socket to form a "bone block door" to protect the roots of adjacent teeth, with a safety distance ≥ 1.5 mm; Bluntly dissect the bone mass and the labial mucoperiosteum, and retain the lingual soft tissue pedicle to preserve the blood supply source.

[0027] In the embodiment of the present invention, an ultrasonic osteotome is used to minimally invasively extract the maxillary and mandibular first premolars while preserving the integrity of the alveolar septum bone wall. This operation can minimize the damage to the alveolar bone and maintain the original structure and stability of the alveolar bone. The intact alveolar septum bone wall can provide a good foundation for subsequent bone healing and repair, facilitating the formation and growth of new bone. The ultrasonic osteotome has a high working frequency and adjustable power, enabling precise cutting during tooth extraction. Compared with traditional tooth extraction methods, it can reduce the traction and damage to surrounding tissues and reduce surgical trauma. At the same time, due to its precise cutting characteristics, it can effectively reduce intraoperative bleeding, making the surgical field clearer and facilitating the doctor's operation. Since the ultrasonic osteotome can quickly and accurately complete the tooth extraction process, the surgical time can be effectively shortened. For patients, this means less pain and fatigue during the operation, and also reduces the risk of infection. Moreover, due to the small surgical trauma and less bleeding, the patient's postoperative recovery time will be significantly shortened.

[0028] In an embodiment of the present invention, a detachable orthognathic osteotomy milling cutter assembly is configured at the distal end of the robotic arm execution module. The orthognathic osteotomy milling cutter assembly includes a rod portion 2, a milling cutter 1, and an exclusion channel. The proximal end of the rod portion 2 is detachably connected to the power output end of the orthognathic surgery robot system; the milling cutter 1 includes a central axis support portion and a plurality of milling portions. The plurality of milling portions are arranged circumferentially on the central axis support portion and are evenly distributed; an exclusion channel is provided between the plurality of milling portions to timely discharge the osteotomy bone fragments, and the maximum radius size of the milling cutter 1 is coupled with the target osteotomy width.

[0029] In an embodiment of the present invention, a detachable orthognathic osteotomy milling cutter assembly is configured at the distal end of the robotic arm execution module. This detachable design enables doctors to flexibly select and replace different types of milling cutter assemblies according to different surgical requirements and patient conditions. For example, for patients with different bone thicknesses or densities, a milling cutter 1 of corresponding specifications can be selected for more precise osteotomy operations. The detachable structure facilitates the cleaning and maintenance of the milling cutter assembly after the operation. Medical staff can remove the milling cutter 1 from the robotic arm and thoroughly clean, disinfect, and inspect it to ensure its performance and safety during the next use. At the same time, if a certain component is damaged or worn, it can also be quickly replaced, reducing the equipment repair time and cost. The milling cutter 1 includes a central axis support portion and a plurality of milling portions. The plurality of milling portions are arranged circumferentially on the central axis support portion and are evenly distributed. This design can achieve more uniform and precise cutting during osteotomy, making the osteotomy surface smoother. Compared with traditional milling cutters 1, it can better control the shape and size of the osteotomy, improve the accuracy of the operation, and reduce damage to surrounding tissues. The evenly distributed milling portions can remove more bone materials in one operation, greatly improving the osteotomy efficiency compared with single-edge or cutter tools with a small number of milling portions. This is particularly important for complex orthognathic surgeries, which can reduce the operation time, lower the patient's anesthesia risk, and surgical trauma.

[0030] In an embodiment of the present invention, the milling portion includes a first extension portion and a second extension portion, and the second extension portion of one milling portion and the first extension portion of the adjacent milling portion form the exclusion channel.

[0031] In the design where the second extension portion of one milling portion and the first extension portion of the adjacent milling portion form an exclusion channel, this combination method cleverly utilizes the structure of the milling portion itself to construct an efficient chip removal path. Compared with the traditional single-channel design or the method of relying on external additional structures for chip removal, this design can more directly and quickly discharge the bone fragments generated during osteotomy. Since the channel is formed by the extension portions of adjacent milling portions, the bone fragments can be quickly sucked into the channel after being generated. This close-range chip removal method helps to reduce the residue of bone fragments in the surgical area, avoids the interference that the accumulation of bone fragments may cause to the surgical operation, enables doctors to see the surgical site more clearly, and thus perform osteotomy operations more accurately.

[0032] In an embodiment of the present invention, the navigation module performs real-time intraoperative registration, combines the three-dimensional anatomical model of the maxillofacial bones with the intraoperative dynamic grating scanning data, and realizes the real-time spatial matching of the bony landmark points to correct the registration drift caused by body position changes.

[0033] In the embodiments of the present invention, the navigation module performs real-time intraoperative registration by combining the three-dimensional anatomical model of the maxillofacial bones and the intraoperative dynamic grating scanning data. The three-dimensional anatomical model provides detailed bone structure information, including the positions and shapes of various bony landmark points, etc.; while the dynamic grating scanning data can reflect the actual anatomical situation during the operation in real time. By fusing the two, more comprehensive and accurate information about the bony landmark points can be obtained, laying a foundation for subsequent precise matching. During the operation, changes in the patient's body position may cause relative changes in the bone position, thereby affecting the accuracy of the operation. This technical solution can promptly detect and correct the registration drift caused by body position changes by performing real-time spatial matching of the bony landmark points. For example, when the patient's head moves slightly during the operation, the system can quickly adjust the registration of the bone model according to the new grating scanning data to ensure that the surgical operation is always based on accurate position information.

[0034] In one embodiment of the present invention, there is also a bone healing assessment model. The input data of the bone healing assessment model includes image data, biological signal data, and mechanical signal data. The fusion module of the bone healing assessment model uses a cross-module attention mechanism to determine the parallel weights of the image data, biological signal data, and mechanical signal data, and generates a fused feature vector; the time axis deviation of different modalities is aligned through dynamic time warping.

[0035] Using a cross-module attention mechanism to determine the parallel weights of the image data, biological signal data, and mechanical signal data can automatically adjust the weights of each data module according to the relative importance of the data for bone healing assessment. By determining appropriate weights and generating a fused feature vector, the most effective information in the multi-source data can be extracted and fused together. This avoids the limitations that may exist in a single data modality and improves the accuracy of the assessment of the degree of bone healing.

[0036] Those skilled in the art can clearly understand that for the sake of convenience and brevity of description, the specific recognition content executed by the above-described system and device can refer to the corresponding process in the foregoing method embodiments.

[0037] The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, provided that these changes fall within the scope of the claims of the present invention and their equivalent technologies, they still fall within the protection scope of the present invention.

Claims

1. An orthognathic surgery robot system, characterized in that, Including: A multimodal data fusion module that integrates cone-beam CT and intraoral scanning. High-resolution three-dimensional bony structure images are obtained through the cone-beam CT to reconstruct a three-dimensional anatomical model of the maxillofacial bones. Sub-millimeter details of the tooth surface and occlusion relationship are captured through the intraoral scanning to construct a functional occlusion model; A navigation module that plans a navigation path based on the three-dimensional anatomical model and functional occlusion model of the target object's maxillofacial bones through a preset adaptive path planning algorithm; A robotic arm execution module that performs removal, osteotomy, and fixation actions based on the navigation path.

2. The orthognathic surgery robot system according to claim 1, characterized in that, The cone-beam CT scans the target object with cone-beam X-rays to generate a three-dimensional model of hard tissues including at least the jawbone and temporomandibular joint, with an accuracy of ±0.2 mm, for pre-operative planning of the osteotomy line position and bone block movement path. The local bone density is quantified through gray value analysis to provide biomechanical parameters for the power adjustment of the ultrasonic bone scalpel to avoid intraoperative bone fracture or insufficient cutting.

3. The orthognathic surgery robot system according to claim 1, characterized in that, The intraoral scanning uses blue / white structured light scanning technology to obtain the surface morphology, adjacency relationship, and dental arch curve of the tooth crown, with a reconstruction accuracy of ±20 μm, for simulating the distribution of postoperative occlusal contact points on a virtual articulator.

4. The orthognathic surgery robot system according to claim 1, characterized in that, The preset adaptive path planning algorithm of the navigation module is the biomechanics-guided artificial potential field method. The construction of the field function of the biomechanics-guided artificial potential field method includes: Attractive force field: Taking the pre-operative planned osteotomy line as the target point, the potential field intensity is positively correlated with the bone density. The formula is: U_att = 0.5k(ρ)⋅||q - q_goal||², k ∈ [50, 200] N / mm, q represents the real-time pose of the end effector of the robotic arm, and q_goal represents the pre-operative planned osteotomy target pose; Repulsive force field: Blood vessels / nerves: Gaussian potential field, peak intensity 500 N / mm², action radius 3 mm; Soft tissue protection: A dynamic potential field that exponentially increases with the deformation amount; The path generation of the biomechanics-guided artificial potential field method includes: The potential field gradient descent method combined with a virtual fixture: Set "non-traversable plane" constraints in sensitive areas and enable the "tunnel guidance" mode for the thick bone area of the mandibular angle.

5. The orthognathic surgery robot system according to claim 1, wherein, The robotic arm execution module's performance of removal, osteotomy, and fixation actions based on the navigation path further includes: Minimally invasive extraction of the maxillary and mandibular first premolars with an ultrasonic bone scalpel, preserving the integrity of the alveolar septal bone wall; Osteotomy at a level 3 - 5 mm below the root apex with a piezoelectric bone scalpel to keep the upper jaw targeted at the nasal floor parallel line and the lower jaw targeted at the lower edge of the mental foramen. The osteotomy depth is the unilateral cortical bone + part of the cancellous bone; Longitudinal osteotomy along the mesial and distal sides of the extraction socket to form a "bone block door", protecting the roots of adjacent teeth, with a safety distance ≥1.5 mm; Bluntly stripping the bone block from the labial mucoperiosteum and preserving the lingual soft tissue pedicle to retain the blood supply source.

6. The orthognathic surgery robot system according to claim 5, wherein, A detachable orthognathic segmented osteotomy milling cutter assembly is configured at the distal end of the robotic arm execution module. The orthognathic segmented osteotomy milling cutter assembly includes a rod portion, a milling cutter, and an exclusion channel. The proximal end of the rod portion is detachably connected to the power output end of the orthognathic surgical robot system; the milling cutter includes a central axis support portion and a plurality of milling portions. The plurality of milling portions are arranged circumferentially on the central axis support portion and are evenly distributed; an exclusion channel is arranged between the plurality of milling portions to timely discharge the osteotomy bone fragments, and the maximum radius dimension of the milling cutter is coupled with the target osteotomy width.

7. The orthognathic surgery robot system according to claim 6, characterized in that, The milling portion includes a first extension portion and a second extension portion, and the second extension portion of one milling portion and the first extension portion of the adjacent milling portion form the exclusion channel.

8. The orthognathic surgery robot system according to claim 1, characterized in that, The navigation module performs real-time intraoperative registration, combines the three-dimensional anatomical model of the maxillofacial bones with the intraoperative dynamic grating scanning data, and realizes the real-time spatial matching of the bony landmark points to correct the registration drift caused by the body position change.

9. The orthognathic surgery robot system according to claim 1, wherein, It further includes a bone healing assessment model. The input data of the bone healing assessment model includes image data, biological signal data, and mechanical signal data. The fusion module of the bone healing assessment model adopts a cross-module attention mechanism to determine the parallel weights of the image data, biological signal data, and mechanical signal data, and generates a fused feature vector. Align the time axis deviation of different modalities through dynamic time warping.

Images