Artificial Intelligence-Based Tooth Arch Matching and Jaw Segmentation Method, System, and Medium

Abstract

The present invention relates to an artificial intelligence-based dentition matching and jaw bone segmentation method, system and medium, including the following steps: S1, constructing a dentition matching algorithm under constrained movement to obtain an objective function for matching a digital dentition model with the dentition of a skull model; S2, minimizing the objective function to obtain an automatic matching method for the upper and lower dentitions with the dentition of the skull model, and the translation and rotation methods of the upper and lower dentitions. The jaw bone segmentation method includes: using this dentition matching method to replace the tooth part on the three-dimensional skull model with the upper and lower dentitions of the digital dentition model; then automatically segmenting the condyles of the three-dimensional skull model to complete the automatic segmentation of the jaw bone of the three-dimensional skull model. This application constructs an intelligent dentition matching method and jaw bone segmentation method for orthognathic surgery, which can automatically complete the important links of jaw bone segmentation and dentition matching in digital orthognathic surgery design by computer, so as to reduce the workload of clinicians and improve work efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of orthodontics, and in particular, to a method, system and medium for tooth alignment matching and jaw segmentation based on artificial intelligence. Background Art

[0002] Orthognathic surgery is the most commonly used method for treating dentofacial deformities, and the precise implementation of the surgery needs to be based on accurate diagnosis and surgical plan design. Traditional orthognathic surgery design mainly relies on detailed clinical examinations, facial image data, two-dimensional plain films, model analysis, etc. for evaluation and design. Although in recent years, with the progress of imaging technology, digital technologies such as computer-aided surgery (CAS) have been applied in orthognathic surgery. The personalized, minimally invasive and precise digital orthognathic surgery design based on CAD / CAM has become the main method, but currently, there is a lack of technical applications for intelligent surgical design of dentofacial deformities at home and abroad, and it is difficult to achieve intelligent and automated diagnosis and treatment design of dentofacial deformities.

[0003] As Figure 1 shown, the digital orthognathic surgery design mainly has the following steps: First is the three-dimensional reconstruction of the skull model, jaw segmentation, tooth alignment matching, surgical design and 3D printing of the corresponding surgical jaw plate or guide plate. Among them, jaw segmentation is because after the three-dimensional reconstruction of the upper and lower dental arches, the condyle and the glenoid fossa are adhesed, and manual separation is required; tooth alignment matching is because the tooth structure is complex, its radiopacity is higher than that of bone tissue, and there will be a large error in the tooth morphology reconstructed from CT data, so it is necessary to replace the jaw dental arch part with a digital dental arch model. At present, the jaw segmentation and tooth alignment matching are manually operated by clinicians, and the process is complicated, the proportion of repetitive labor is high, the efficiency is low, and the time consumption is long.

[0004] In recent years, with the increasing close combination of artificial intelligence and medicine, digital orthognathic surgery has gradually moved towards artificial intelligence orthognathic surgery. If automatic segmentation of the upper and lower jaws and tooth alignment matching can be achieved, it will effectively reduce the workload of clinicians and save working time. Summary of the Invention

[0005] The present application provides a method and system for jaw segmentation and tooth alignment matching based on artificial intelligence to solve the above technical problems.

[0006] The present application is achieved through the following technical solutions:

[0007] The method for tooth alignment matching based on artificial intelligence provided by the present application includes the following steps:

[0008] S1, constructing a tooth alignment matching algorithm under constrained motion to obtain an objective function for matching the digital tooth alignment model and the tooth alignment of the skull model ;

[0009]

[0010] In the above formula, is the upper dentition node of the digital dentition model, is the lower dentition node of the digital dentition model, is the relative sliding axis between the upper and lower dentitions, is the relative sliding amount between the upper and lower dentitions, and are both nodes on the skull, and are the overall rigid body rotation and translation;

[0011] S2, minimize the objective function The obtained optimal solution is the automatic matching method of the upper and lower dentitions with the dentition of the skull model, and the translation and rotation methods of the upper and lower dentitions.

[0012] Among them, before the step S1, there is also a preparation work, and the preparation work includes:

[0013] Collect maxillofacial spiral CT data;

[0014] Collect the plaster bite model corresponding to the patient when taking the maxillofacial spiral CT, and record the patient's bite relationship;

[0015] Obtain a digital dentition model, including:

[0016] Step 1, scan the upper and lower dentitions respectively to obtain the complete upper and lower dentitions;

[0017] Step 2, position the upper and lower dentition models in the best bite with bite registration wax, and at the same time scan the upper and lower models to obtain the relative relationship between the upper and lower dentitions;

[0018] Step 3, match the upper and lower dentitions obtained in Step 1 with the relative relationship in Step 2 in the software of the scanning device. After the scanning is completed, export the upper and lower dentition models that record the relative position relationship.

[0019] The jaw bone segmentation method based on artificial intelligence provided by this application includes the following steps:

[0020] Reconstruct a three-dimensional skull model based on maxillofacial spiral CT data;

[0021] Adopt a dentition matching method to replace the tooth part on the three-dimensional skull model with the upper and lower dentitions of the digital dentition model;

[0022] Segment the condyle of the three-dimensional skull model, including:

[0023] Precise positioning of the condyle to automatically find the position of the condyle;

[0024] Condyle clipping. Based on the surface normal features of the condyle and the condylar fossa, the surface point sets of the condyle and the condylar fossa are classified, and unnecessary points are trimmed according to the two types of point sets, realizing the segmentation of the condyle.

[0025] Specifically, the method for precise positioning of the condyle is as follows:

[0026] Taking the condyle obtained by manually segmenting the upper and lower jaws of a skull model as the reference model, and the condyle of another skull model that needs to be automatically segmented as the target model. Let the nodes of the target model be , and the nodes of the reference model be , and the matching objective function is expressed as :

[0027]

[0028] In the above formula, and are the overall rotation and translation, is the change in the model scale. Minimize the objective function , and the obtained optimal solution is the node for positioning the condyle.

[0029] Specifically, the method for condyle clipping is as follows:

[0030] Use to represent the surface point set of the condyle:

[0031]

[0032] In the above formula, is the normal vector at the node . Use N to represent the local area around the condyle, is the center point coordinate of the area N, is the unit vector in the vertical direction;

[0033] Use to represent the surface point set of the condylar fossa:

[0034]

[0035] According to and Automatically trim off unnecessary points to realize the segmentation of the condyle.

[0036] It should be noted that since the bone threshold of the condylar part is close to the standard bone threshold, and the bone threshold of the condylar fossa part is close to the cortical bone threshold, when performing condylar segmentation, it is necessary to first perform threshold reconstruction to roughly identify the range of the condyle and the condylar fossa through threshold reconstruction, so as to achieve the purpose of preliminary accurate positioning of the condyle; then perform subsequent accurate positioning of the condyle.

[0037] The dentition matching and jaw bone segmentation system based on artificial intelligence provided by the present application includes:

[0038] A data import module for importing a three-dimensional skull model and a digital dentition model;

[0039] A dentition matching and skull bone segmentation module, which uses the jaw bone segmentation method to replace the tooth part on the three-dimensional skull model with the maxillary dentition and mandibular dentition of the digital dentition model and segment the condylar part of the three-dimensional skull model.

[0040] An apparatus provided by the present application includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned jaw bone segmentation method based on artificial intelligence is implemented.

[0041] The present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above-mentioned jaw bone segmentation method based on artificial intelligence is implemented.

[0042] Compared with the prior art, the present application has the following beneficial effects:

[0043] Based on the concept of artificial intelligence, the present application constructs an intelligent dentition matching method and jaw bone segmentation method for orthognathic surgery, which can realize the important link of jaw bone segmentation and dentition matching in digital orthognathic surgery design by computer, so as to reduce the workload of clinicians and improve work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the embodiments of the present application, and constitute a part of the present application, but do not constitute a limitation to the embodiments of the present invention.

[0045] Figure 1 is a flowchart of digital orthognathic surgery design;

[0046] Figure 2 In (a) are schematic diagrams of nodes of the upper and lower jaw dentitions of the jaw bone, and in (b) are schematic diagrams of nodes of the digital upper and lower jaw dentitions;

[0047] Figure 3 is a normal vector diagram of the area around and the surface of the condyle in the embodiment;

[0048] Figure 4 It is a schematic diagram of the construction program in the embodiment; among them, (a) is the spiral CT data and plaster dental cast model collected as the training set, (b) is a schematic diagram of the upper and lower dental arch nodes on the jaw bone and the digital upper and lower dental arch nodes before matching, (c) is a schematic diagram of the upper and lower dental arch nodes on the jaw bone and the digital upper and lower dental arch nodes after matching, (d) is the well-matched upper and lower dental arches in the system, (e) is a schematic diagram of the preliminary positioning of the condyle, (f) is a schematic diagram of the precise positioning of the condyle, and (g) is the automatically segmented condyle;

[0049] Figure 5 It is a schematic diagram of automatic dental arch matching in the embodiment;

[0050] Figure 6 It is a schematic diagram of automatic jaw bone segmentation in the embodiment;

[0051] Figure 7 It is a schematic diagram of the overall deviation between automatic segmentation and manual segmentation of the maxilla in the embodiment;

[0052] Figure 8 It is a statistical analysis chart of the overall deviation between automatic segmentation and manual segmentation of the maxilla in the embodiment;

[0053] Figure 9 It is a schematic diagram of the overall deviation between automatic segmentation and manual segmentation of the mandible in the embodiment;

[0054] Figure 10 It is a statistical analysis chart of the overall deviation between automatic segmentation and manual segmentation of the mandible in the embodiment;

[0055] Figure 11 It is a comparison chart of the working time between automatic segmentation and manual segmentation for skeletal asymmetric malocclusion in clinical practice in the embodiment;

[0056] Figure 12 It is a comparison chart of the working time between automatic segmentation and manual segmentation for class II malocclusion in clinical practice in the embodiment;

[0057] Figure 13 It is a comparison chart of the working time between automatic segmentation and manual segmentation for class III malocclusion in clinical practice in the embodiment. Specific implementation manners

[0058] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations.

[0059] Accordingly, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0060] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other. It should be noted that the various embodiments in this specification are described in a progressive manner, and the key point of each embodiment is to illustrate the differences from other embodiments. For the parts that are the same or similar among the various embodiments, reference can be made to each other. For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the corresponding parts of the method embodiments for the relevant parts.

[0061] The method for dentition matching and jaw segmentation based on artificial intelligence disclosed in this embodiment includes the following steps:

[0062] 1. Preliminary preparation work

[0063] First, collect all maxillofacial spiral CT data. This step can be performed using a PNMS MX16 EVO (Philips, pitch: 7.25 mm, slice thickness: 1 mm, slice interval: 0.5 mm, photographing range: 768 mm * 768 mm, scanning time: 26.2 s, tube voltage: 120 KV, tube current: 282 mA, voxel: 0.33 mm3) for scanning to obtain medical digital imaging and communication format data.

[0064] Then, obtain all plaster dentition models: collect the corresponding plaster bite models when the patient undergoes maxillofacial spiral CT scanning, that is, dentition models. The production process of the bite model: Use silicone rubber impression material to make impressions of the patient's upper and lower dentitions, and obtain the patient's dentition plaster model by pouring with high-strength dental stone, and record the patient's occlusion relationship with occlusion registration wax.

[0065] Obtain a digital dentition model: Transform the plaster dentition model into data that can be recognized by a computer through scanning. Specifically, it includes: (1) Scan the maxillary dentition to obtain a complete maxillary dentition; (2) Scan the mandibular dentition to obtain a complete mandibular dentition; (3) Use occlusion registration wax to position the upper and lower dentition models in the best occlusion, and fix them using scanning accessories. At the same time, scan the upper and lower dentition models to obtain the relative relationship between the upper and lower dentitions; (4) Match the upper and lower dentitions in the first two steps with the relative relationship in step (3) in the software of the scanning device. After scanning, the upper and lower dentition models with the recorded relative position relationship are exported in STL format. This step can be performed using a chamber-type optical 3D scanner for scanning.

[0066] It should be noted that collecting maxillofacial spiral CT data and plaster dental models can be used as a training set. By performing commonality analysis on a certain number of CT models and digital dental models, an algorithm for dental arch matching and segmentation can be obtained.

[0067] Perform three-dimensional reconstruction on the craniofacial bone tissue in the CT data, and at the same time scan to obtain a digital dental model and construct a data set. This data set can be used for accuracy experiments, including comparison of cusp landmark points, the overall dental arch, and clinical working time, etc.

[0068] In some embodiments, the training set includes maxillofacial spiral CTs of multiple patients with skeletal Class II, Class III, and asymmetric deformities and the corresponding plaster dental models.

[0069] , construct an automated dental arch matching algorithm

[0070] The positional constraint between the upper dental arch and the lower dental arch can be simplified to a relative sliding along a known axis, and the upper and lower dental arches as a whole are matched with the cranial dental arch through a rigid body transformation. Assume that the imported upper dental arch nodes are , and the imported lower dental arch nodes are ， And the relative sliding axis between the upper and lower dental arches is , and the relative sliding amount is , then the matching of the imported dental arch and the cranial dental arch can be expressed as the objective function , as shown in Equation (1):

[0071] (1)

[0072] In Equation (1), and are all nodes on the skull, and are the overall rigid body rotation and translation. is the upper dental arch node of the skull model, has a matching correspondence; is the lower dental arch node of the skull model, and has a matching correspondence; m refers to the number of imported digital upper dental arch nodes , and n refers to the number of imported digital lower dental arch nodes . Minimizing this objective function can be reduced to an unconstrained quadratic optimization problem, and this equation has a standard solution process. The obtained optimal solution is the automatic matching method of the upper and lower dental arches with the skull model, and the translation and rotation methods of the upper and lower dental arches.

[0073] By constructing a dentition matching algorithm under constrained motion, the upper and lower dentitions can be matched simultaneously, and the preliminary positional relationship between the two can be obtained, such as Figure 2 as shown

[0074] , construct an automatic jaw segmentation algorithm

[0075] In this implementation, automatic segmentation of the upper and lower jaws is established: there are two places that need to be processed in the establishment of this part. One is the segmentation of the dentition part, and the other is the segmentation of the joint area.

[0076] 2.1 Segmentation of the dentition part

[0077] After the dentition is matched in the previous stage, since the upper and lower dentitions are scanned separately, the tooth parts on the skull model are replaced by the upper and lower dentitions, and the segmentation of the dentition part is completed. In order to further divide the skull model into the upper and lower jaws, it is necessary to segment the condyle and the dentition part.

[0078] 2.2 Segmentation of the condyle part.

[0079] After the dentition is segmented, the condyle needs to be segmented. Since the bone threshold of the condyle part is close to the standard bone threshold, and the bone threshold of the condyle fossa part is close to the cortical bone threshold, when segmenting the condyle, it is necessary to first perform threshold reconstruction to roughly identify the range of the condyle and the condyle fossa through threshold reconstruction, so as to achieve the purpose of preliminary condyle fine positioning; then perform subsequent condyle fine positioning. The segmentation of the condyle part specifically includes the following steps

[0080] S2.1, preliminary positioning: establish different threshold models: including after obtaining the required CT data and the corresponding STL dentition, reconstruct the CT according to the following three thresholds. The first is the standard bone threshold (standard bone model); the second is the cortical bone threshold (cortical bone model); the third is the dental threshold (dental model).

[0081] S2.2, condyle fine positioning: Automatically finding the position of the condyle is the prerequisite for segmenting the condyle. First, use the condyle obtained by manually segmenting the upper and lower jaws of a skull model as the reference model, and use the condyle of the skull model that needs to be automatically segmented as the target model. Let the nodes of the target model be , and the nodes of the reference model be , and the matching objective function is expressed as , as shown in Equation (2):

[0082] (2)

[0083] In Equation (2), and are the overall rotation and translation, while is the change in the model scale. For this objective function The minimization can be reduced to an unconstrained quadratic optimization problem, which has a standard solution process, and the obtained optimal solution is the node for locating the condyle.

[0084] S2.3, Condyle clipping:

[0085] The condyle clipping algorithm is one of the keys to realizing condyle segmentation. After locating the condyle, the processing of the condyle clipping algorithm is limited to a local area around the condyle. In this area, the surface normal information has significant features, such as Figure 3 shown.

[0086] Figure 3 In, represents the normal vector at a point on the condyle surface, represents the normal vector at a point on the fossa surface of the condyle; its feature is that the normal vectors on the condyle surface basically point to the outside of the local area, while the normal vectors on the fossa surface of the condyle all point to the inside of the area and tend to point downward. This feature can be used to classify the condyle and the fossa of the condyle. Let N represent the local area around the condyle, and represent the set of condyle surface points in it, then can be defined as:

[0087] (3)

[0088] where is the normal vector at node , is the central point coordinate of region N, is the unit vector in the vertical direction. The set realizes the classification of condyle surface points, and similarly, the classification of condyle surface points can be realized; the set of fossa surface points is , can be defined as:

[0089] (4)

[0090] By clipping off unnecessary points according to these two types of point sets, the segmentation of the condyle is realized.

[0091] Based on the above method, the present application discloses an artificial intelligence-based dentition matching and jaw segmentation system, which can be used to implement the above artificial intelligence-based dentition matching and jaw segmentation method, and specifically includes:

[0092] A data import module for respectively importing a three-dimensional skull model and a digital dentition model;

[0093] The dentition matching and skull segmentation module uses the jawbone segmentation method to replace the tooth part on the three-dimensional skull model with the upper and lower dentition of the digital dentition model and segment the condyle part of the three-dimensional skull model.

[0094] According to the above method, in this embodiment, an orthognathic surgery intelligent design program (hereinafter referred to as IDOS) was constructed, as Figure 4 shown. The program interface was developed using the C / C++ language, based on the C++11 standard. Specifically, the Qt graphical interface library was used to implement C / C++ programming and achieve the OpenGL three-dimensional graphics rendering function. It can run across platforms and supports Windows 7 and Windows 10. The imported three-dimensional skull model and digital upper and lower dentition models can be imported into the program respectively, and the computer automatically completes dentition matching and the segmentation of the upper and lower jaws (condyles), as Figure 5 、 Figure 6 shown.

[0095] In this embodiment, 320 cases of patients with skeletal Class II, Class III, and asymmetric deformities, including their maxillofacial spiral CT and corresponding plaster dentition models, were selected as the training set for constructing the algorithm. After successful construction, another 140 cases were selected for testing to verify the effectiveness of the algorithm. The verification is as follows: In the experiment to test the accuracy of IDOS for automatic dentition matching, the maxillofacial spiral CT data of 140 cases of dentofacial deformities who had undergone orthognathic surgery and the corresponding digital dentition models in the same period were selected. IDOS was used to automatically complete jawbone segmentation and dentition matching, while clinicians manually completed jawbone segmentation and dentition matching, and the differences in the dental cusp landmark points and the differences in the overall upper and lower dentition obtained by the two methods were analyzed. The experimental results are shown in Table 1 and Figures 7 - 10 , where Figure 8 and Figure 10 from left to right are: average deviation distance, average positive deviation, average negative deviation, and standard deviation.

[0096] Table 1 Precision test experiment: differences in dental cusp landmark points

[0097]

[0098] Landmark description: A is the subspinale point, U1(R) is the mesioincisal angle point of the right maxillary central incisor, U3(R / L) is the cusp point of the right / left maxillary canine, U6(R / L) is the mesiobuccal cusp point of the right / left maxillary first molar, B is the supramentale point, L1(R) is the mesioincisal angle point of the right mandibular central incisor, L3(R / L) is the cusp point of the right / left mandibular canine, and L6(R / L) is the mesiobuccal cusp point of the right / left mandibular first molar.

[0099] The three-dimensional spatial positions of these 12 cusp points were compared. They were decomposed into three directions of X, Y, and Z for further experimental testing. Excluding the three landmark points of U6 (R / L) and L6 (R), there were no statistical differences in the remaining landmark points. At the same time, the color atlas of the upper and lower dental arches as a whole showed that the overall deviation was within 2 mm, indicating that the results of automatically matching the dental arches using IDOS were relatively accurate.

[0100] The clinical working time spent on manually segmenting the jaws and matching the dental arches and the clinical working time spent on automated jaw segmentation and dental arch matching were recorded separately. According to the results, it was shown that the clinical working time using intelligent jaw segmentation and dental arch matching was significantly shortened, from an average of 40 minutes to an average of 2 minutes, greatly reducing the workload of clinicians, as Figures 11 - 13 shown.

[0101] In several embodiments provided by this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of this application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0102] In addition, in each embodiment of this application, the various functional modules can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0103] When the above functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs. It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the element.

[0104] The above specific implementation manners further elaborate in detail the purpose, technical solution and beneficial effects of this application. It should be understood that the above is only the specific implementation manners of the present invention and is not used to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An artificial intelligence-based dental arch matching method, characterized in that: It includes the following steps: S1. Construct a dentition matching algorithm under constrained movement to obtain an objective function for the dentition that matches the digital dentition model and the cranial model ; In the above formula, is the upper dentition node of the digital dental arch model, is the lower dentition node of the digital dental arch model, is the relative sliding axis between the upper and lower dentitions, is the relative sliding amount between the upper and lower dentitions, is the upper dentition node of the skull model, has a matching correspondence; is the lower dentition node of the skull model, and has a matching correspondence; and are the overall rigid body rotation and translation; The number of digitized upper dentition nodes imported by m The number of digitized lower dentition nodes imported by n The number; S2, minimize the objective function to obtain the automatic matching method between the upper and lower dental arches and the dental arch of the skull model, as well as the translation and rotation methods of the upper and lower dental arches; Before the step S1, it also includes preparatory work, which includes: Collecting maxillofacial spiral CT data; Collecting the corresponding plaster occlusion model when the patient takes the maxillofacial spiral CT and recording the patient's occlusion relationship; Obtaining a digital dentition model, including: Step 1, scanning the maxillary dentition and the mandibular dentition respectively to obtain the complete maxillary dentition and mandibular dentition; Step 2, positioning the upper and lower dentition models in the best occlusion with occlusion recording wax, and at the same time scanning the upper and lower dentition models to obtain the relative relationship between the upper and lower dentitions; Step 3, matching the maxillary and mandibular dentitions obtained in Step 1 with the relative relationship in Step 2 in the software of the scanning device. After the scanning is completed, export the maxillary and mandibular dentition models recording the relative position relationship.

2. Artificial intelligence-based jaw segmentation method, characterized in that: It includes the following steps: Reconstructing a three-dimensional skull model based on the maxillofacial spiral CT data; Adopting the dentition matching method as described in Claim 1, and replacing the tooth part on the three-dimensional skull model with the maxillary dentition and the mandibular dentition of the digital dentition model; Automatically segmenting the condyle part of the three-dimensional skull model, including: accurate condyle positioning, automatically finding the position of the condyle; Condyle clipping, based on the surface normal features of the condyle and the condyle fossa, classifying the surface point sets of the condyle and the condyle fossa, and trimming unnecessary points according to the two types of point sets, realizing the automatic segmentation of the condyle.

3. The jawbone segmentation method according to claim 2, characterized in that: The method for accurate condyle positioning is: Taking the condyle obtained by manually splitting the upper and lower jaws of a skull model as the reference model, and the condyle of another skull model that needs to be automatically segmented as the target model. Let the nodes of the target model be , and the nodes of the reference model be . The matching objective function is expressed as : In the above formula, and are the overall rotation and translation, is the change in the model scale. By minimizing the objective function the nodes for locating the condyle are obtained.

4. The jawbone segmentation method according to claim 2 or 3, characterized in that: The method for condyle clipping is: Use to represent the point set on the condylar surface: In the above formula, is the normal vector at the node Let N denote the local area around the condyle, is the central point coordinate of the region N, is the unit vector in the vertical direction; Use to represent the set of surface points of the condylar fossa: According to and Automatically trim unnecessary points to achieve the segmentation of the condyle.

5. The jawbone segmentation method according to claim 2 or 3, characterized in that: Before the accurate condyle positioning, it also includes the step of initial condyle positioning, and the initial condyle positioning includes: establishing different threshold models, and roughly identifying the range of the condyle and the condyle fossa through threshold reconstruction.

6. The jawbone segmentation method according to claim 5, wherein: The threshold reconstruction specifically is: after obtaining the required CT data and the corresponding STL dentition, reconstructing the CT data according to the following three thresholds: The first one, the standard bone threshold; The second one, the cortical bone threshold; The third one, the dental threshold.

7. An artificial intelligence-based dental arch matching and jaw segmentation system, characterized in that: It includes: A data import module for importing the three-dimensional skull model and the digital dentition model; A dentition matching and skull segmentation module, adopting the jaw segmentation method as described in any one of Claims 2-6, replacing the tooth part on the three-dimensional skull model with the maxillary dentition and the mandibular dentition of the digital dentition model and segmenting the condyle part of the three-dimensional skull model.

8. An artificial intelligence-based dentition matching device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it realizes the method as described in any one of Claims 2-6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it realizes the method as described in any one of Claims 2-6.

Images