A method and system for evaluating the accuracy of oral implantation

Abstract

The application provides an oral implant precision evaluation method and system, and the evaluation method comprises the following steps: acquiring preoperative and postoperative images, the preoperative image is used to acquire the planned geometric characteristics of the implant, and the planned geometric characteristics comprise a planned center; target transformation matrix is acquired by performing a first scheme or a second scheme according to the actual situation of the postoperative image to register the preoperative and postoperative images, and a target implant region is acquired on the postoperative image; the actual geometric characteristics of the implant are acquired based on the target implant region; and the oral implant precision is evaluated according to the planned geometric characteristics, the actual geometric characteristics and the target transformation matrix. The evaluation method considers the actual situation of the postoperative image to select a registration scheme and an implant recognition scheme, so that the evaluation method has the advantage of high accuracy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of medical treatment, and particularly relates to a method and system for evaluating the precision of oral implantation. BACKGROUND

[0002] A dental implant is a device that is surgically implanted into the maxilla or mandible of a human body at a toothless site, and a false tooth is installed on the upper part of the dental implant after the surgical wound is healed. In other words, the dental implant is equivalent to the tooth root of a real tooth.

[0003] The implantation position of a dental implant is very important, and improper implantation may damage the adjacent anatomical structures such as the mandibular nerve canal and the maxillary sinus. Therefore, the implantation precision of a dental implant plays an important role in the development and application of oral implantation technology. Therefore, it is very important to develop an implantation precision evaluation method for a dental implant.

[0004] Most of the implantation precision evaluation methods for dental implants in the prior art include registering the preoperative image and the postoperative image of the patient, extracting the dental implant image on the postoperative image, obtaining the actual root tip point and the actual implant point of the dental implant based on the extracted dental implant image, comparing the actual root tip point with the planned root tip point based on the registration result of the preoperative image and the postoperative image, and comparing the actual implant point with the planned implant point to determine the implant position deviation and the angle deviation. When these implantation precision evaluation methods are executed, there is an implicit assumption that the dental implant is a complete circle in the image slice of the postoperative image. However, in actual situations, the dental implant may not be a complete circle or a complete circle due to various reasons such as the angle and the type of the dental implant, that is, the actual situation does not meet the above implicit assumption, which leads to inaccurate evaluation results. SUMMARY

[0005] The purpose of the present application is to provide an oral implantation precision evaluation method and system, which aims to reasonably select some steps in the oral implantation precision evaluation method according to the actual situation of the implant region, and avoid the situation that the extraction result is inaccurate due to the shape of the implant not meeting the assumed condition.

[0006] To achieve the above-mentioned purpose, the present application provides an oral implantation precision evaluation method, comprising:

[0007] obtaining a preoperative image and a postoperative image of a target object, the preoperative image being used to obtain a planned implantation position of an implant, so as to obtain a planned geometric feature of the implant, the planned geometric feature comprising a planned center;

[0008] performing a first scheme to obtain a first transformation matrix between the preoperative image and the postoperative image, and identifying a first implant region on the postoperative image;

[0009] determining whether the ellipticity of the first implant region meets a predetermined requirement, if yes, taking the first transformation matrix as a target transformation matrix and taking the first implant region as a target implant region, if no, executing a second scheme to obtain a second transformation matrix between the preoperative image and the postoperative image, and identifying a second implant region on the postoperative image, and taking the second transformation matrix as the target transformation matrix and taking the second implant region as the target implant region;

[0010] obtaining an actual geometric feature of the implant based on the target implant region;

[0011] evaluating the oral implant precision according to the planned geometric feature, the actual geometric feature and the target transformation matrix;

[0012] One of the first scheme and the second scheme comprises a first registration scheme and a first identification scheme, and the other of the first scheme and the second scheme comprises a second registration scheme and a second identification scheme; the first registration scheme is different from the second registration scheme, and the first identification scheme is different from the second identification scheme.

[0013] Optionally, the first registration scheme comprises: performing primary registration on the preoperative image and the postoperative image based on a nearest iteration method to obtain a first primary transformation matrix between the preoperative image and the postoperative image; and performing secondary registration on the preoperative image and the postoperative image based on the first primary transformation matrix to obtain a corresponding transformation matrix; and the first identification scheme comprises: transforming the planned center to the postoperative image based on the first transformation matrix to obtain a first calculated center of the implant; and performing a region growing algorithm in a first threshold range with the first calculated center as a seed point to obtain a corresponding implant region.

[0014] The second registration scheme comprises: performing primary registration on the preoperative image and the postoperative image based on a mutual information metric to obtain a second primary transformation matrix; and performing secondary registration on the preoperative image and the postoperative image based on the second primary transformation matrix to obtain a corresponding transformation matrix; and the second identification scheme comprises: selecting at least one first region of interest in the postoperative image, each of the first regions of interest comprising at least one implant; marking pixel points or voxel points within a second threshold range in each of the first regions of interest; and discriminating a pixel point set or a voxel point set formed by the marked pixel points or voxel points according to a length feature of the implant, and screening out the pixel point set or the voxel point set meeting the length feature as a corresponding implant region.

[0015] Optionally, when the first scheme comprises the first registration scheme and the first identification scheme, the predetermined condition comprises that an ellipticity of the first implant region is greater than or equal to a first preset threshold; when the first scheme comprises the second registration scheme and the second identification scheme, the predetermined condition comprises that the ellipticity of the first implant region is less than the first preset threshold.

[0016] Optionally, the step of performing initial registration on the preoperative image and the postoperative image based on the Levenberg-Marquardt method comprises:

[0017] Performing isosurface extraction on the preoperative image and the postoperative image based on a second preset threshold, and obtaining a preoperative reconstruction model and a postoperative reconstruction model;

[0018] According to the implantation site of the implant, selecting at least one second region of interest on the preoperative reconstruction model, and selecting at least one third region of interest on the postoperative reconstruction model, the third region of interest corresponding to the second region of interest one by one;

[0019] Obtaining a part of the preoperative reconstruction model located in the second region of interest, and obtaining a part of the postoperative reconstruction model located in the third region of interest;

[0020] Using visibility discrimination, obtaining a first outermost curved surface of the part of the preoperative reconstruction model located in the second region of interest, and obtaining a first point set constituting the first outermost curved surface, and obtaining a second outermost curved surface of the part of the postoperative reconstruction model located in the third region of interest, and obtaining a second point set constituting the second outermost curved surface;

[0021] Registering the first point set and the second point set based on the Levenberg-Marquardt method to obtain the first primary transformation matrix.

[0022] Optionally, the step of performing initial registration on the preoperative image and the postoperative image based on the mutual information metric comprises:

[0023] According to the implantation site of the implant, selecting a fourth region of interest in the preoperative image, and selecting a fifth region of interest in the postoperative image, the fifth region of interest corresponding to the fourth region of interest;

[0024] Obtaining a part of the preoperative image located in the fourth region of interest, and obtaining a part of the postoperative image located in the fifth region of interest;

[0025] The portions of the preoperative image located in the fourth region of interest and the portions of the postoperative image located in the fifth region of interest are registered based on a mutual information metric to obtain a second primary transformation matrix.

[0026] Optionally, the step of performing secondary registration on the preoperative image and the postoperative image comprises:

[0027] The postoperative image is corrected based on the first primary transformation matrix or the second primary transformation matrix to obtain a postoperative corrected image.

[0028] At least one sixth region of interest is selected in the preoperative image, each of the sixth regions of interest comprising a planned implant position, and at least one seventh region of interest is selected in the postoperative corrected image, each of the seventh regions of interest comprising the implant, the seventh regions of interest corresponding to the sixth regions of interest one by one.

[0029] The corresponding sixth regions of interest and the corresponding seventh regions of interest are registered based on a mutual information metric to obtain the transformation matrix.

[0030] Optionally, the target implant region is saved as a binary image; the actual geometric features comprise an actual center, an actual apex point, an actual implant point and an actual axis direction of the implant.

[0031] The step of obtaining the actual geometric features of the implant based on the target implant region is selectively performed according to a first calculation scheme or a second calculation scheme.

[0032] The first calculation scheme comprises:

[0033] A second moment of the binary image is obtained, and at least one equivalent axis is obtained based on the second moment of the binary image, and an extension direction of the equivalent axis with the largest length is taken as the actual axis direction of the implant.

[0034] A centroid of the binary image is obtained, and the centroid of the binary image is taken as the actual center of the implant.

[0035] An actual axis of the implant is constructed, the actual axis passing through the actual center and extending along the actual axis direction of the implant.

[0036] An axial bounding box of the binary image is obtained, and two intersection points of the axial bounding box and the actual axis of the implant are obtained to be respectively taken as the actual implant point and the actual apex point.

[0037] The second calculation scheme comprises:

[0038] obtaining a second moment of the binary image to obtain at least one equivalent axis, and taking an extension direction of the equivalent axis with the largest length as an estimated axis direction of the implant;

[0039] obtaining a center of mass of the binary image, and taking the center of mass of the binary image as an actual center of the implant;

[0040] optimizing the estimated axis direction according to a predetermined step to obtain an actual axis direction;

[0041] constructing an actual axis of the implant, the actual axis passing through the actual center and extending along the actual axis direction;

[0042] projecting the binary image along a predetermined direction to obtain a first projection region, and taking two intersection points of the first projection region and the actual post-conchoid as the actual implant point and the actual apex point respectively; the predetermined direction is perpendicular to the actual axis.

[0043] Optionally, the predetermined step comprises repeatedly adjusting the estimated axis direction, and after each adjustment, projecting the binary image along the current estimated axis direction to obtain a second projection region; comparing all the second projection regions, screening out the second projection region with the smallest area, and taking the estimated axis direction corresponding to the second projection region with the smallest area as the actual axis direction.

[0044] Optionally, the actual geometric features comprise an actual center, an actual apex point, an actual implant point and an actual axis direction; the planning geometric features further comprise a planning apex point, a planning implant point and a planning axis direction.

[0045] The step of evaluating the oral implant precision according to the planning geometric features, the actual geometric features and the transformation matrix comprises:

[0046] transforming the actual geometric features to the preoperative image based on the target transformation matrix to obtain calculated geometric features, the calculated geometric features comprising a second calculated center corresponding to the actual center, a calculated implant point corresponding to the actual implant point, a calculated apex point corresponding to the actual apex point, and a calculated axis direction corresponding to the actual axis direction;

[0047] constructing a calculated axis of the implant, the calculated axis passing through the second calculated center and extending along the calculated axis direction;

[0048] Construct a first positioning line and a second positioning line, and use the intersection of the first positioning line and the calculation axis as the reference implantation point, and the intersection of the second positioning line and the calculation axis as the reference root apex point; the first positioning line passes through the calculation implantation point and is perpendicular to the calculation axis, and the second positioning line passes through the calculation root apex point and is perpendicular to the calculation axis;

[0049] The deviation between the reference implantation point and the planned implantation point, the deviation between the reference root tip point and the planned root tip point, and the angle between the actual axial direction and the planned axial direction are obtained.

[0050] To achieve the above objectives, the present invention also provides an oral implant accuracy evaluation system, including a data processing workstation configured to perform the oral implant accuracy evaluation method as described above.

[0051] Optionally, the oral implant precision evaluation system also includes a standard model, which is used to provide standard preoperative images and standard postoperative images;

[0052] The oral implant precision evaluation system is configured to perform the oral implant precision evaluation method based on the preoperative and postoperative images of the standard model, in order to assess the accuracy of the oral implant precision evaluation method.

[0053] Optionally, the standard model includes a matrix, a planting base, and markers. The planting base is embedded in the matrix and has a reserved planting position. The planting position is exposed outside the matrix. There are multiple markers, and the multiple markers are asymmetrically arranged on the matrix.

[0054] The standard preoperative image refers to the image of the implantation site of the standard model before the hole is prepared, and the standard postoperative image refers to the image of the implantation site of the standard model after the implant is placed.

[0055] Optionally, the standard model includes a 3D printed model based on preoperative images of the target object. The 3D printed model includes an initial 3D printed model and a planned 3D printed model. The planned 3D model has reserved implantation positions, and the location of the implantation positions is determined according to the preoperative planning scheme.

[0056] The standard preoperative image refers to the image of the initial 3D printed model, and the standard postoperative image refers to the image of the implant or filling material at the implantation site of the planned 3D printed model.

[0057] Compared with existing technologies, the oral implant accuracy evaluation method and system of the present invention have the following advantages:

[0058] The aforementioned method for evaluating the accuracy of oral implantation includes: acquiring preoperative and postoperative images of the target patient, wherein the preoperative images are used to obtain the planned implantation location of the implant, and to obtain the planned geometric features of the implant, the planned geometric features including the planning center; executing a first scheme to obtain a first transformation matrix between the preoperative and postoperative images, and identifying a first implant region on the postoperative image; determining whether the ellipticity of the first implant region meets a predetermined requirement, if yes, then using the first transformation matrix as the target transformation matrix and the first implant region as the target implant region, if no, then executing a second scheme to obtain a second transformation matrix between the preoperative and postoperative images, and to determine whether the ellipticity of the first implant region meets a predetermined requirement. The method involves identifying a second implant region on the postoperative image, using the second transformation matrix as the target transformation matrix and the second implant region as the target implant region; acquiring the actual geometric features of the implant based on the target implant region; and evaluating the accuracy of the oral implantation based on the planned geometric features, the actual geometric features, and the target transformation matrix. One of the first and second schemes includes a first registration scheme and a first identification scheme, while the other includes a second registration scheme and a second identification scheme. The first registration scheme differs from the second registration scheme, and the first identification scheme differs from the second identification scheme. In this oral implantation accuracy evaluation method, the initial registration scheme and implant region acquisition scheme are determined by judging whether the ellipticity of the acquired implant region meets the requirements. This avoids the problem of inaccurate extracted implant regions due to the implant's shape in the image slice not meeting the set conditions, thereby preventing inaccurate evaluation results in subsequent acquisitions. Attached Figure Description

[0059] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein:

[0060] Figure 1 This is a schematic diagram illustrating an application scenario of the oral implant precision evaluation system provided by the present invention according to an embodiment;

[0061] Figure 2 This is an overall flowchart of the oral implant precision evaluation method provided by the present invention according to an embodiment;

[0062] Figure 3 This is a flowchart of the first identification scheme for obtaining the implant region in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0063] Figure 4 This is a flowchart of the second identification scheme for obtaining the implant region in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0064] Figure 5 This is a flowchart of the registration of preoperative and postoperative images in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0065] Figure 6 This is a flowchart of the first registration scheme for preoperative and postoperative images in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0066] Figure 7 This is a schematic diagram of the oral implant accuracy evaluation method provided by the present invention according to an embodiment, in the process of registering preoperative and postoperative images, based on the discriminability of processing the part of the preoperative reconstructed model located in the second region of interest, wherein a) shows the visibility discrimination processing before processing and b) shows the visibility discrimination processing after processing.

[0067] Figure 8 This is a flowchart illustrating the second registration scheme for preoperative and postoperative images in the oral implant accuracy evaluation method provided by an embodiment of the present invention.

[0068] Figure 9 This is a schematic diagram of the mutual information measurement method used in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0069] Figure 10 This is a flowchart of the registration of preoperative and postoperative images in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0070] Figure 11 This is a flowchart of the process of executing a first calculation scheme to obtain actual geometric features in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0071] Figure 12 This is a flowchart illustrating the second calculation scheme for obtaining actual geometric features in the oral implant accuracy evaluation method provided by an embodiment of the present invention;

[0072] Figure 13 This is a schematic diagram illustrating the planned implantation point, planned apical point, planned axial direction, reference implantation point, reference apical point, and calculated axial direction during the execution of the oral implantation accuracy evaluation method provided by the present invention according to an embodiment.

[0073] Figure 14 This is a flowchart of the oral implant accuracy evaluation method provided by an embodiment of the present invention, which evaluates the accuracy of oral implantation based on planned geometric features, actual geometric features, and the transformation matrix between preoperative and postoperative images.

[0074] Figure 15This is a block diagram of the data processing workstation of the oral implant accuracy evaluation system provided by the present invention according to an embodiment;

[0075] Figure 16 This is a schematic diagram of the standard model in different states in the oral implant accuracy evaluation system provided by an embodiment of the present invention. The standard model in the figure includes a matrix, an implant base and markers. The standard model shown in a) is in the first state, the standard model shown in b) is in the second state, and the first standard model shown in c) is in the third state. Only some markers are shown in the figure.

[0076] Figure 17 This is a schematic diagram of a standard model in the oral implant accuracy evaluation system provided by the present invention according to an embodiment. In the diagram, a) is the initial 3D printed model, b) is the planned 3D printed model, and c) is the planned 3D printed model when the implant is placed or the filling material is filled at the implantation site. Detailed Implementation

[0077] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show components related to the present invention and are not drawn according to the actual number, shape, and size of components in the actual implementation. In the actual implementation, the type, quantity, and proportion of each component can be arbitrarily changed, and the component layout may also be more complex.

[0078] Furthermore, while each embodiment described below possesses one or more technical features, this does not imply that users of the present invention must simultaneously implement all technical features in any embodiment, or can only separately implement some or all technical features in different embodiments. In other words, provided it is feasible, those skilled in the art can, based on the disclosure of the present invention and depending on design specifications or implementation requirements, selectively implement some or all technical features in any embodiment, or selectively implement a combination of some or all technical features in multiple embodiments, thereby increasing the flexibility in implementing the present invention.

[0079] The purpose of this invention is to provide an oral implant precision evaluation system, which can be used to perform an oral implant precision evaluation method. This oral implant evaluation method does not rely on the shape of the implant in the image slice, but can extract the implant area from the postoperative image and obtain the actual geometric features of the implant. This avoids the problem of low accuracy of the obtained implant area image due to the implant shape not meeting the requirements in the image slice, thereby improving the accuracy of oral implant precision evaluation.

[0080] To make the objectives, advantages, and features of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. It should be noted that the drawings are all in a very simplified form and use non-precise proportions, and are only used to facilitate and clearly illustrate the objectives of the embodiments of the present invention. The same or similar reference numerals in the drawings represent the same or similar parts.

[0081] Figure 1 This diagram illustrates an application scenario of the oral implant precision evaluation system 1 provided in this embodiment of the invention. The oral implant precision evaluation system 1 is applied to oral implant surgery to assist in determining the accuracy of the implant placement. Oral implant surgery includes steps such as preoperative planning, implant placement, abutment installation, permanent abutment installation, and placement of the implant crown. In the preoperative planning stage, the dentist plans the implant placement based on the patient's preoperative CBCT images. In the implant placement stage, such as... Figure 1 As shown, based on the navigation system 200, the dentist can complete the procedures of preparing the cavity and inserting the implant with the assistance of the robotic arm 100. After the implant is placed, the dentist can take postoperative CBCT images of the patient. The oral implant precision evaluation system 1 provided in this embodiment of the invention can evaluate the implant placement precision based on the patient's preoperative and postoperative CBCT images.

[0082] The dental implant precision evaluation system 1 may include a data processing workstation 10 (e.g., Figure 15 As shown in the figure, the data processing workstation 10 is used to perform oral implant accuracy evaluation methods based on the patient's preoperative and postoperative images to obtain the corresponding patient's oral implant accuracy results. Figure 2 The flowchart illustrates the dental implant accuracy evaluation method executed by data processing workstation 10. (Example) Figure 2 As shown, this method for evaluating the accuracy of dental implants includes the following steps:

[0083] Step S100: Acquire preoperative and postoperative images of the target object. The preoperative images are used to plan the implant placement location to obtain the planned implant placement location, and then obtain the planned geometric features of the implant, including the planning center. All images mentioned in this article refer to CBCT images.

[0084] Step S200: Register the preoperative and postoperative images to obtain the target transformation matrix between the preoperative and postoperative images. Also, obtain the target implant region on the postoperative image.

[0085] Step S300: Obtain the actual geometric features of the plant based on the target plant region.

[0086] Step S400: Evaluate the accuracy of dental implantation based on the planned geometric features, actual geometric features, and the target transformation matrix between preoperative and postoperative images.

[0087] In this embodiment of the invention, step S200 may specifically include: executing a first scheme to obtain a first transformation matrix between preoperative and postoperative images, and identifying a first implant region on the postoperative image. Then, determining whether the ellipticity of the first implant region meets predetermined requirements; if yes, using the first transformation matrix as the target transformation matrix and the first implant region as the target implant region; if no, executing a second scheme to obtain a second transformation matrix between preoperative and postoperative images, identifying a second implant region on the postoperative image, and using the second transformation matrix as the target transformation matrix and the second implant region as the target implant region.

[0088] The first scheme and the second scheme each include a first registration scheme and a first identification scheme, and the other scheme includes a second registration scheme and a second identification scheme. The first registration scheme is different from the second registration scheme, and the first identification scheme is different from the second identification scheme.

[0089] In other words, during step S200, the ellipticity of the implant region is used to determine whether the transformation matrix between the acquired implant region and the preoperative and postoperative images is appropriate, thereby eliminating situations where the shape of the implant in the postoperative image slice does not conform to the implicit assumption. In other words, the oral implant accuracy evaluation method provided in this embodiment selects appropriate registration and implant identification schemes based on the actual characteristics of the postoperative images, thereby improving the accuracy of the registration results and the acquired implant region, resulting in a higher accuracy of the final acquired oral implant precision.

[0090] Specifically, the first registration scheme includes performing initial registration on preoperative and postoperative images based on the most recent iteration method to obtain a first primary transformation matrix between the preoperative and postoperative images. Then, a second registration is performed on the preoperative and postoperative images based on the first primary transformation matrix to obtain a corresponding transformation matrix (i.e., a first transformation matrix or a second transformation matrix) between the preoperative and postoperative images.

[0091] Figure 3 A flowchart of the first identification scheme is shown. (Example) Figure 3As shown, the first identification scheme includes:

[0092] Step S210: Based on the transformation matrix between preoperative and postoperative images, the planning center is transformed to the postoperative image to obtain the first calculation center of the implant on the postoperative image.

[0093] Step S220: Set a first threshold range. The first threshold range may include the CT value of the implant in preoperative and / or postoperative images.

[0094] Step S230: On the postoperative image, the first calculation center is used as the seed point, and the region growth algorithm is executed based on the first threshold range, and the calculation result is used as an implant region.

[0095] It should be understood that each implant has a planned planting location and corresponding planned geometric features. Therefore, the aforementioned planning center refers to the planning center of each implant, and the first calculation center refers to the first calculation center of each implant. Each implant region includes one implant. When multiple implants are included, step S310 includes converting all planning centers to the postoperative image to obtain multiple first calculation centers. Step S330 then includes performing a region growing algorithm using each first calculation center as a seed point to obtain multiple implant regions.

[0096] The second registration scheme includes performing initial registration on preoperative and postoperative images based on mutual information metric to obtain a second primary transformation matrix between the preoperative and postoperative images. Then, performing secondary registration on the preoperative and postoperative images based on the second primary transformation matrix to obtain corresponding transformation matrices (i.e., a first transformation matrix or a second transformation matrix).

[0097] Figure 4 A flowchart of the second identification scheme is shown. (Example) Figure 4 As shown, the second identification scheme includes:

[0098] Step S210': Select at least one first region of interest on the postoperative image, each first region of interest including at least one implant.

[0099] Step S220': Set a second threshold range. The second threshold range may include the CT value of the implant in preoperative and / or postoperative images.

[0100] Step S230': Mark the pixels or voxels located within the second threshold range in each first region of interest.

[0101] Step S240': Based on the length characteristics of the implant, the set of pixels formed by the marked pixels or the set of voxels formed by the marked voxels are judged, and the set of pixels or voxels that meet the length characteristics of the implant are selected, and the set of pixels or voxels that meet the length characteristics of the implant is used as an implant region.

[0102] Taking the example of pixels marked in step S330', the number of pixel sets formed by all marked pixels within each first region of interest is equal to the number of implants included in the corresponding first region of interest. That is, if each first region of interest includes n implants, then all marked pixels within that first region of interest form a set of n pixels, where n is a positive integer greater than or equal to 1. Thus, each implant region includes one implant. Those skilled in the art know how to determine the pixel set, so it will not be elaborated here. Step S340' actually includes determining the longest axis as the principal axis in the corresponding pixel set based on the distribution of pixels in each pixel set, and then determining whether the length of the principal axis of each pixel set matches the length of the implant. If so, the corresponding pixel set is considered to conform to the length characteristics of the implant. Alternatively, when there are multiple pixel sets, step S340' makes a judgment by combining the ratio of the lengths of the principal axes in all first regions of interest with the length of the implant.

[0103] When the first scheme includes a first registration scheme and a first identification scheme, the transformation matrix obtained through the first registration scheme is the first transformation matrix, and the implant region obtained through the first identification scheme is the first implant region. The aforementioned predetermined condition refers to the ellipticity of the first implant region being greater than or equal to a first preset threshold. When the first scheme includes a second registration scheme and a second identification scheme, the transformation matrix obtained through the second registration scheme is the first transformation matrix, and the implant region obtained through the second identification scheme is the first implant region. The aforementioned predetermined condition refers to the ellipticity of the first implant region being less than the first preset threshold.

[0104] It should be understood that obtaining the ellipticity of the plant region is something that is known to those skilled in the art, and therefore will not be elaborated here.

[0105] The following section will provide a more detailed explanation of the methods for evaluating the precision of dental implants.

[0106] During step S100, the data processing workstation 10 can acquire preoperative and postoperative images wirelessly via a wireless communication module. Alternatively, the doctor can manually input the preoperative and postoperative images into the data processing workstation 10. Acquiring the preoperative images allows for the determination of the planned implantation location, and consequently, the acquisition of the planned geometric features.

[0107] As previously described, the registration of preoperative and postoperative images includes steps S250 and S260 (e.g. Figure 5 (As shown). Step S250 includes initial registration of the preoperative and postoperative images based on the most recent iteration method or mutual information metric method, and obtaining a primary transformation matrix (first primary transformation matrix or second primary transformation matrix) between the preoperative and postoperative images. Step S260 includes secondary registration of the preoperative and postoperative images based on the primary transformation matrix, specifically including steps S261 and S262. Step S261 includes correcting the postoperative image based on the primary transformation matrix to obtain a corrected postoperative image. Step S262 includes secondary registration of the corrected postoperative image and the preoperative image to obtain a transformation matrix between the preoperative and postoperative images. That is, in this embodiment of the invention, the preoperative and postoperative images are first initially aligned by executing step S250, and then precisely aligned by executing steps S261 and S262, thereby improving the registration accuracy of the preoperative and postoperative images and improving the accuracy of subsequent oral implant evaluation. Those skilled in the art will know that by performing a registration operation, preoperative and postoperative images are unified to the same coordinates, thereby enabling the planned geometric features to be compared with the actual geometric features.

[0108] Figure 6 The flowchart illustrates the initial registration of preoperative and postoperative images based on the most recent iteration method. Figure 6 As shown, the steps for initial registration of preoperative and postoperative images based on the most recent iteration method include:

[0109] Step S251: Based on a second preset threshold, isosurfaces are extracted from the preoperative and postoperative images to obtain a preoperative reconstruction model and a postoperative reconstruction model. The second preset threshold is determined based on the preoperative and / or postoperative images, and may include the CT value of the implant in the preoperative and / or postoperative images.

[0110] Step S252: Based on the implant site, select at least one second region of interest on the preoperative reconstruction model. And, select at least one third region of interest on the postoperative reconstruction model. In practice, the implant site includes at least one of the patient's maxilla and mandible. The second region of interest may include several teeth near the planned implant site, for example, including 2-3 teeth to the left of the implant site or 2-3 teeth to the right of the implant site. The third region of interest corresponds to the second region of interest.

[0111] Step S253: Process the preoperative reconstruction model by removing the portion of the preoperative reconstruction model located outside the second region of interest, and retaining the portion of the preoperative reconstruction model located within the second region of interest. This portion is referred to as the first model portion 20 (e.g., Figure 7 (as shown in a)). Furthermore, the postoperative reconstruction model is processed to remove the portion of the postoperative reconstruction model located outside the third region of interest, retaining the portion located within the third region of interest; this portion is referred to as the second model part.

[0112] Step S254: Process the first model part 1 based on the visibility discrimination method to obtain the outermost surface of the first model part 1. This outermost surface is referred to as the first outermost surface 30 (e.g., Figure 7 (As shown). Then, the first point set constituting the first outermost surface 30 is obtained. Similarly, the second model part is processed based on the visibility discrimination method to obtain the outermost surface of the second model part, which can be called the second outermost surface, and then the second point set constituting the second outermost surface is obtained.

[0113] Step S255: Register the first point set and the second point set based on the Iterative Closest Point (ICP) and obtain the first primary transformation matrix.

[0114] Because teeth are rigidly connected to the jawbone, local registration of the first and second model parts achieves the registration of the preoperative remodel and the postoperative reconstructed model. Local registration significantly reduces the computational load, increases calculation speed, and lowers the software configuration requirements for the oral implant accuracy evaluation system.

[0115] Furthermore, the reason for executing step S254 is that the first model part 20 and the second model part still have certain internal structures and cannot be directly used for ICP registration. By selecting points outside the corresponding reconstructed model region as the viewpoint to observe the reconstructed model, the outermost surface of the corresponding reconstructed model can be obtained, and then the point set constituting the corresponding outermost surface can be obtained for ICP registration.

[0116] Figure 8 This diagram illustrates a flowchart of initial registration of preoperative and postoperative images based on mutual information measurement. Figure 8 As shown, the steps for initial registration of preoperative and postoperative images based on mutual information measurement include:

[0117] Step S251': Based on the implantation site (i.e., the patient's maxilla and / or mandible), select the fourth region of interest in the preoperative images and the fifth region of interest in the postoperative images. Similar to the first registration scheme, the fourth region of interest may include several teeth near the planned implantation site, such as 2-3 adjacent teeth on the left or right side of the implantation site. The fifth region of interest corresponds one-to-one with the fourth region of interest.

[0118] Step S252': Remove the portion of the preoperative image located outside the fourth region of interest, and retain the portion of the preoperative image located within the fourth region of interest; similarly, remove the portion of the postoperative image located outside the fifth region of interest, and retain the portion of the preoperative image located within the fifth region of interest. This reduces the amount of data involved in subsequent registration steps, thereby reducing computational load, increasing computational speed, and lowering the software configuration requirements of the dental implant accuracy evaluation system.

[0119] Step S253': Based on mutual information, the portion of the preoperative image located within the fourth region of interest and the portion of the postoperative image located within the fifth region of interest are registered to obtain the second primary transformation matrix. Those skilled in the art will understand that mutual information measurement is an iterative optimization process, the principle of which is as follows... Figure 9 As shown, the portion of the preoperative image located within the fourth region of interest is used as a fixed image 01, and a point set is acquired. The portion of the postoperative image located within the fifth region of interest is used as a floating image 02, and a point set is acquired. The point set of the floating image 02 is applied to interpolator 04, and mutual information is measured between it and the point set of the fixed image 01. The mutual information measure is used as the iterative input to optimizer 03, and the transform parameter set is updated. The updated transform parameter set is repeatedly applied to the floating image 02 and interpolator 04, and resampling is performed to update the mutual information measure. Each updated measure is used as the input data for the next iteration of optimizer 03. This process continues until the convergence condition of optimizer 03 is met.

[0120] Step S261 is performed in accordance with existing technology and will not be described in detail here.

[0121] Step S262 can be referred to Figure 10 The steps shown are to be performed, including:

[0122] Step S262a: Select at least one sixth region of interest in the preoperative images, each sixth region of interest including a planned implantation site. And select at least one seventh region of interest in the postoperative images, each seventh region of interest including an implant, with each seventh region of interest corresponding to a sixth region of interest. Preferably, the number of both sixth and seventh regions of interest is one.

[0123] Step S262b: Based on mutual information measurement, register the portion of the preoperative image located within the sixth region of interest and the portion of the postoperative image located within the seventh region of interest to obtain the transformation matrix between the preoperative and postoperative images. In this step, the portion of the postoperative image located within the seventh region of interest is used as a floating image to perform mutual information measurement.

[0124] Furthermore, it should be noted that although both steps S253' and S262b use mutual information measurement to perform the registration operation, the optimization parameters differ because the required registration accuracy varies between the two steps. For step S253', only approximate alignment of the preoperative and postoperative images is required; therefore, a more aggressive optimization strategy should be adopted. In step S262b, precise alignment of the preoperative and postoperative images is desired; therefore, a more robust optimization strategy should be adopted. The specific settings are well-known to those skilled in the art and will not be elaborated here.

[0125] The actual geometric features of the implant include the actual center, actual root apex, actual implantation point, and actual axial direction. Each target implant region acquired in step S200 is saved as a binary image. Step S300 is selectively executed according to either the first calculation scheme or the second calculation scheme.

[0126] Figure 11 A flowchart of the first calculation scheme is shown. (For example...) Figure 11 As shown, the first calculation scheme includes:

[0127] Step S310: Obtain the second moment of the binary image, and obtain at least one equivalent axis based on the second moment of the binary image, with the extension direction of the longest equivalent axis being taken as the actual axis direction of the implant. It should be understood that when there is only one equivalent axis, the extension direction of that equivalent axis is the actual axis direction. Obtaining the equivalent axis based on the second moment of the binary image is well known to those skilled in the art and will not be elaborated here.

[0128] Step S320: Obtain the centroid of the binary image and use the centroid of the binary image as the actual center of the implant.

[0129] Step S330: Construct the actual axis of the plant based on the actual axis direction and the actual center of the plant. The actual axis passes through the actual center and extends along the extension direction of the plant.

[0130] Step S340: Obtain the axial bounding box of the binary image, and obtain the two intersection points between the axial bounding box and the actual axis of the implant. The two intersection points are respectively used as the actual implantation point and the actual root tip point of the implant.

[0131] Figure 12 A flowchart of the second calculation scheme is shown. (For example...) Figure 12 As shown, the second calculation scheme includes:

[0132] Step S310': Obtain the second moment of the binary image, and obtain at least one equivalent axis based on the second moment of the binary image, and take the extension direction of the equivalent axis with the largest length as the estimated axis direction of the implant. It should be understood that when there is only one equivalent axis, the extension direction of the equivalent axis is the estimated axis direction.

[0133] Step S320': Optimize the estimated axis direction according to the predetermined steps and obtain the actual axis direction.

[0134] Step S330': Obtain the centroid of the binary image and use the centroid of the binary image as the actual center of the implant.

[0135] Step S340': Based on the actual axis direction and the actual center, construct the actual axis of the implant. The actual axis passes through the actual center and extends along the actual axis direction.

[0136] Step S350': Project the binary image along a predetermined direction to obtain the first projection region. The predetermined direction is perpendicular to the actual axis.

[0137] Step S360': Obtain two intersection points between the first projection area and the actual axis, which will be used as the actual implantation point and the actual root apex point, respectively.

[0138] The predetermined step in step S330' refers to repeatedly adjusting the estimated axis direction and, after each adjustment, projecting the binary image along the current estimated axis direction to obtain a second projection region. All second projection regions are compared, and the region with the smallest area is selected. The estimated axis direction corresponding to the second projection region with the smallest area is taken as the actual axis direction. Here, the step of adjusting the axis direction may include: constructing a straight line extending along the initial estimated axis direction, and then changing the estimated axis direction by rotating the straight line. The initial estimated axis direction refers to the estimated axis direction obtained through step S310'.

[0139] The accuracy of the actual apical point, implantation point, and axial direction of the implant obtained by executing the second calculation scheme is higher than that obtained by executing the first calculation scheme. Furthermore, regardless of whether the first or second calculation scheme is used to obtain the actual geometric features, it does not depend on the shape of the implant in the image slice, thus avoiding the problem of reduced accuracy of the obtained actual geometric features due to the implant's shape in the image slice not meeting the requirements. In other words, both the first and second calculation schemes provided in this embodiment of the invention are beneficial for improving the accuracy of oral implant precision evaluation.

[0140] The aforementioned planning geometric features also include Figure 13 The planned implantation point 41, planned root tip point 42, and planned axis direction of the plant are shown. The planned axis direction is shown as arrow S1. Figure 14 The flowchart of step S400 is shown, as follows: Figure 14 As shown, step S400 specifically includes the following steps: S410, S420, S430, S440 and S450.

[0141] Step S410 includes transforming the actual geometric features onto the preoperative image based on the target transformation matrix between the preoperative and postoperative images, obtaining the computational geometric features of the implant on the preoperative image. The computational geometric features include a second computational center corresponding to the actual center, a computational implantation point corresponding to the actual implantation point, a computational apex point corresponding to the actual apex point, and a computational axis direction corresponding to the actual axis, as shown in the figure. Figure 13 As shown by arrow S2.

[0142] Step S420 includes constructing the computational axis of the implant based on the second computational center and the direction of the computational axis. The computational axis passes through the second computational center and extends along the direction of the computational axis.

[0143] Step S430 includes constructing a first positioning line and a second positioning line. The first positioning line passes through the calculated implantation point and is perpendicular to the calculation axis. The second positioning line passes through the calculated root apex point and is perpendicular to the calculation axis.

[0144] Step S440 includes obtaining the intersection point of the first positioning line and the calculation axis as the reference implantation point 43 (e.g. Figure 13 As shown), and obtain the intersection of the second positioning line and the calculation axis as the reference apex point 44 (as shown). Figure 13 (As shown).

[0145] Step S450 includes obtaining the deviation between the reference implantation point and the planned implantation point, the deviation between the reference root apex point and the planned root apex point, and calculating the angle between the axial direction and the planned axial direction. Those skilled in the art will understand that the deviation between the reference implantation point and the planned implantation point, the deviation between the reference root apex point and the planned root apex point, and the angle between the calculated axial direction and the planned axial direction can characterize the implantation accuracy.

[0146] Based on the above introduction to the methods for evaluating the accuracy of dental implants, the data processing workstation 10 includes, as follows: Figure 15 The diagram shows a data acquisition module 300, a registration module 400, an implant identification module 500, a judgment module 600, a geometric feature calculation module 700, and a precision calculation module 800. The data acquisition module 300 is configured to execute step S100. The registration module 400 is communicatively connected to the data acquisition module 300. The judgment module 600 is communicatively connected to the registration module 400 and the implant identification module, and is used to determine whether the ellipticity of the implant region meets a predetermined condition; that is, the registration module 400, the implant identification module 500, and the judgment module 600 are jointly used to execute step S200. The geometric feature calculation module 700 is communicatively connected to the implant identification module 500 and the judgment module, and is configured to execute step S300. The precision calculation module 800 is communicatively connected to the registration module 400 and the geometric feature calculation module 700, and is configured to execute step S400.

[0147] More specifically, the registration module 400 includes a first initial registration module 410, a second initial registration module 420, and a second registration module 430. The first initial registration module 410 and the second initial registration module 420 are independent of each other, and are communicatively connected to the second registration module 430. The first initial registration module 410 is configured to execute a first registration scheme, the second initial registration module 420 is configured to execute a second registration scheme, and the second registration module 430 is configured to execute steps S220 and S230.

[0148] The implant identification module 500 includes a first implant identification module 510 and a second implant identification module 520 that are independent of each other. The first implant identification module 510 is communicatively connected to the registration module 400 and is configured to execute a first identification scheme. The second implant identification module 520 is configured to execute a second identification scheme.

[0149] The geometric feature calculation module 700 includes a first geometric feature calculation module 710 and a second geometric feature calculation module 720 that are independent of each other. The first geometric feature calculation module 710 is configured to execute a first calculation scheme, and the second geometric feature calculation module 720 is configured to execute a second calculation scheme.

[0150] Furthermore, such as Figure 16 and Figure 17 As shown, the oral implant accuracy evaluation system also includes a standard model 50, which provides standard preoperative and postoperative images. When the oral implant accuracy evaluation system 1 performs the oral implant accuracy evaluation method based on the standard preoperative and postoperative images, the accuracy of the oral implant accuracy evaluation method itself can be assessed based on the obtained evaluation results.

[0151] In one optional implementation, such as Figure 16 As shown, the standard model 50 includes a substrate 51, an implant base 52, and markers 53. The substrate 51 is made of, for example, resin and can have a cuboid structure. The implant base 52 includes an extracted maxilla and / or mandible of an animal; optional animals include small pigs or dogs. The markers 53 are made of radiopaque material. An implantation site is pre-drilled on the implant base 52, which is embedded in the substrate 51, and the implantation site is exposed outside the substrate 51 without being obscured by it. Multiple markers 53 are asymmetrically arranged on the substrate 51.

[0152] The standard model 50 has three states: state one, state two, and state three. In state one, the implantation site is empty; no implant is placed. In state two, an implant and connecting rod are placed on the standard model. In state three, an implant is placed on the standard model, but the connecting rod is removed.

[0153] CBCT was used to scan the standard model in its first state, obtaining an image of the standard model in this state, which can be used as a standard preoperative image. CBCT was then used to scan the standard model in its third state, obtaining an image of the standard model in this state, which can be used as a standard postoperative image.

[0154] Furthermore, for this type of standard model, doctors can obtain the planning geometric features using the following methods:

[0155] Step S01: Use a coordinate measuring machine to measure the coordinates of each marker point 53 on the standard model in the first state and the coordinates of the planting position.

[0156] Step S02: Obtain the coordinates of each marker point on the standard preoperative model. The specific implementation of this step is well known to those skilled in the art and will not be elaborated here.

[0157] Step S03: Based on the coordinates of the marker point 53 obtained by the coordinate measuring machine and the coordinates of the marker points on the standard preoperative model, obtain the coordinate system transformation relationship between the standard model and the standard preoperative image.

[0158] Step S04: Based on the coordinate system transformation relationship between the standard model and the standard preoperative image, transform the coordinates of the implantation position on the standard model to the standard preoperative image to serve as the planning position of the standard model, thereby obtaining the planning geometric features.

[0159] In another implementation, such as Figure 17 As shown, the standard model ( Figure 17 (Not marked in the text) includes a 3D printed model obtained from the patient's preoperative images, specifically including an initial 3D printed model 54 and a planned 3D printed model 55, on which a pre-reserved implantation position is reserved, which is determined according to the preoperative planning scheme.

[0160] Thus, the steps for obtaining a standard model, as well as standard preoperative and standard postoperative images, include:

[0161] Step S001: Perform three-dimensional reconstruction based on the patient's preoperative images to obtain a preoperative virtual three-dimensional model.

[0162] Step S002: 3D print the preoperative virtual 3D model to obtain the initial 3D printed model 54 (e.g., Figure 14 (As shown).

[0163] Step S003: Scan the initial 3D printed model 54 using CBCT to obtain an image of the initial 3D printed model 54, which will serve as a standard preoperative image.

[0164] Step S004: Perform preoperative planning based on the patient's preoperative images to determine the planned implantation location, thereby obtaining the planned geometric features of the implant.

[0165] Step S005: Print the preoperative virtual 3D model based on the preoperative planning to obtain the planned 3D printing model 55. Leave the area corresponding to the planned planting position in the planned 3D printing model 55 blank as the planting position.

[0166] Step S005: Insert the implant or filler material (such as...) into the planting position of the planned 3D printed model 55. Figure 17 (As shown in the black part of c), the material of the filler is a developing material.

[0167] Step S006: Use CBCT to scan the planned 3D printed model 55 of the implantation site to obtain an image of the planned 3D printed model 55 in its current state, as a standard postoperative image.

[0168] While the present invention has been disclosed above, it is not limited thereto. Those skilled in the art can make various modifications and variations to the present invention without departing from its spirit and scope. Therefore, if such modifications and variations fall within the scope of the claims and their equivalents, the present invention also intends to include such modifications and variations.

Claims

1. A method for evaluating the accuracy of dental implants, characterized in that, include: Acquire preoperative and postoperative images of the target object. The preoperative images are used to obtain the planned implantation location of the implant to obtain the planned geometric features of the implant, including the planning center. The first scheme is executed to obtain a first transformation matrix between the preoperative image and the postoperative image, and to identify a first implantation region on the postoperative image; Determine whether the ellipticity of the first implant region meets the predetermined requirements. If yes, use the first transformation matrix as the target transformation matrix and the first implant region as the target implant region. If no, execute the second scheme to obtain the second transformation matrix between the preoperative image and the postoperative image, and identify the second implant region on the postoperative image, using the second transformation matrix as the target transformation matrix and the second implant region as the target implant region. The actual geometric features of the implant are obtained based on the target implant region; The accuracy of the dental implant is evaluated based on the planned geometric features, the actual geometric features, and the target transformation matrix. Wherein, one of the first scheme and the second scheme includes a first registration scheme and a first identification scheme, and the other of the first scheme and the second scheme includes a second registration scheme and a second identification scheme; the first registration scheme is different from the second registration scheme, and the first identification scheme is different from the second identification scheme; The first registration scheme includes: performing initial registration on the preoperative image and the postoperative image based on the most recent iteration method to obtain a first primary transformation matrix between the preoperative image and the postoperative image; performing secondary registration on the preoperative image and the postoperative image based on the first primary transformation matrix to obtain a corresponding transformation matrix; the first identification scheme includes: transforming the planning center to the postoperative image based on the first transformation matrix to obtain a first calculation center of the implant; using the first calculation center as a seed point within a first threshold range to perform a region growing algorithm to obtain a corresponding implant region; The second registration scheme includes: performing initial registration on the preoperative image and the postoperative image based on mutual information metric to obtain a second primary transformation matrix; performing secondary registration on the preoperative image and the postoperative image based on the second primary transformation matrix to obtain a corresponding transformation matrix; the second identification scheme includes: selecting at least one first region of interest in the postoperative image, each first region of interest including at least one implant; marking pixels or voxels within a second threshold range in each first region of interest; judging the set of pixels or voxels formed by the marked pixels according to the length characteristics of the implant, and selecting the set of pixels or voxels that meets the length characteristics as the corresponding implant region.

2. The method for evaluating the accuracy of dental implants according to claim 1, characterized in that, When the first scheme includes the first registration scheme and the first identification scheme, the predetermined requirement includes that the ellipticity of the first implant region is greater than or equal to a first preset threshold; when the first scheme includes the second registration scheme and the second identification scheme, the predetermined requirement includes that the ellipticity of the first implant region is less than the first preset threshold.

3. The method for evaluating the accuracy of dental implants according to claim 1, characterized in that, The step of performing initial registration of the preoperative and postoperative images based on the most recent iteration method includes: Based on a second preset threshold, isosurfaces are extracted from the preoperative and postoperative images to obtain preoperative and postoperative reconstruction models. Based on the implantation site, at least one second region of interest is selected on the preoperative reconstruction model, and at least one third region of interest is selected on the postoperative reconstruction model, wherein the third region of interest corresponds one-to-one with the second region of interest; Obtain the portion of the preoperative reconstruction model located within the second region of interest, and obtain the portion of the postoperative reconstruction model located within the third region of interest; Using visibility discrimination, the first outermost surface of the portion of the preoperative reconstruction model located within the second region of interest is obtained, and the first point set constituting the first outermost surface is obtained; and the second outermost surface of the portion of the postoperative reconstruction model located within the third region of interest is obtained, and the second point set constituting the second outermost surface is obtained. The first point set and the second point set are registered using the most recent iteration method to obtain the first primary transformation matrix.

4. The method for evaluating the accuracy of dental implants according to claim 1, characterized in that, The step of performing initial registration of the preoperative and postoperative images based on mutual information measurement includes: Based on the implantation site, a fourth region of interest is selected in the preoperative image, and a fifth region of interest is selected in the postoperative image, wherein the fifth region of interest corresponds to the fourth region of interest. Acquire the portion of the preoperative image located within the fourth region of interest, and acquire the portion of the postoperative image located within the fifth region of interest; Based on mutual information measurement, the portion of the preoperative image located in the fourth region of interest and the portion of the postoperative image located in the fifth region of interest are registered to obtain the second primary transformation matrix.

5. The method for evaluating the accuracy of dental implants according to claim 1, characterized in that, The steps for secondary registration of the preoperative and postoperative images include: The postoperative image is corrected based on the first primary transformation matrix or the second primary transformation matrix to obtain a postoperative corrected image. At least one sixth region of interest is selected in the preoperative images, each sixth region of interest including a planned implantation site; and at least one seventh region of interest is selected in the postoperative corrective images, each seventh region of interest including one implant, and the seventh region of interest corresponds one to the sixth region of interest. Based on mutual information metric, the corresponding sixth region of interest and seventh region of interest are registered to obtain the transformation matrix.

6. The method for evaluating the accuracy of dental implants according to claim 1, characterized in that, The target implant region is saved as a binary image; the actual geometric features include the actual center of the implant, the actual root apex, the actual implantation point, and the actual axial direction; The step of obtaining the actual geometric features of the implant based on the target implant region is selectively performed according to either the first calculation scheme or the second calculation scheme. The first calculation scheme includes: The second-order distance of the binary image is obtained, and at least one equivalent axis is obtained based on the second-order distance of the binary image. The extension direction of the equivalent axis with the largest length is taken as the actual axis direction of the implant. Obtain the centroid of the binary image, and use the centroid of the binary image as the actual center of the implant; Construct the actual axis of the plant, which passes through the actual center and extends along the actual axis direction of the plant; Obtain the axial bounding box of the binary image, and obtain two intersection points of the axial bounding box with the actual axis of the implant, which are respectively used as the actual implantation point and the actual root tip point; The second calculation scheme includes: The second-order distance of the binary image is obtained to obtain at least one equivalent axis, and the extension direction of the equivalent axis with the largest length is used as the estimated axis direction of the implant. Obtain the centroid of the binary image, and use the centroid of the binary image as the actual center of the implant; The estimated axis direction is optimized according to predetermined steps to obtain the actual axis direction; Construct the actual axis of the plant, which passes through the actual center and extends along the direction of the actual axis; The binary image is projected along a predetermined direction to obtain a first projection area, and the two intersections of the first projection area and the actual axis are respectively used as the actual implantation point and the actual root apex point; the predetermined direction is perpendicular to the actual axis.

7. The method for evaluating the accuracy of dental implants according to claim 6, characterized in that, The predetermined steps include: repeatedly adjusting the estimated axis direction, and after each adjustment, projecting the binary image along the current estimated axis direction to obtain a second projection region; comparing all the second projection regions, selecting the second projection region with the smallest area, and using the estimated axis direction corresponding to the second projection region with the smallest area as the actual axis direction.

8. The method for evaluating the accuracy of dental implants according to claim 1, characterized in that, The actual geometric features include the actual center, actual root apex, actual implantation point, and actual axial direction; the planned geometric features also include the planned root apex, planned implantation point, and planned axial direction. The step of evaluating the accuracy of dental implantation based on the planned geometric features, the actual geometric features, and the transformation matrix includes: Based on the target transformation matrix, the actual geometric features are transformed to the preoperative image to obtain computational geometric features. The computational geometric features include a second computational center corresponding to the actual center, a computational implantation point corresponding to the actual implantation point, a computational root apex point corresponding to the actual root apex point, and a computational axis direction corresponding to the actual axis direction. The computational axis of the implant is constructed, the computational axis passes through the second computational center and extends along the direction of the computational axis; Construct a first positioning line and a second positioning line, and use the intersection of the first positioning line and the calculation axis as the reference implantation point, and the intersection of the second positioning line and the calculation axis as the reference root apex point; the first positioning line passes through the calculation implantation point and is perpendicular to the calculation axis, and the second positioning line passes through the calculation root apex point and is perpendicular to the calculation axis; The deviation between the reference implantation point and the planned implantation point, the deviation between the reference root tip point and the planned root tip point, and the angle between the actual axial direction and the planned axial direction are obtained.

9. A dental implant precision evaluation system, characterized in that, Includes a data processing workstation configured to perform the oral implant accuracy evaluation method as described in any one of claims 1-8.

10. The oral implant precision evaluation system according to claim 9, characterized in that, The oral implant precision evaluation system also includes a standard model, which is used to provide standard preoperative images and standard postoperative images; The oral implant precision evaluation system is configured to perform the oral implant precision evaluation method based on the preoperative and postoperative images of the standard model, in order to assess the accuracy of the oral implant precision evaluation method.

11. The oral implant precision evaluation system according to claim 10, characterized in that, The standard model includes a matrix, a planting base, and markers. The planting base is embedded in the matrix and has a reserved planting position. The planting position is exposed outside the matrix. There are multiple markers, which are asymmetrically arranged on the matrix. The standard preoperative image refers to the image of the implantation site of the standard model before the hole is prepared, and the standard postoperative image refers to the image of the implantation site of the standard model after the implant is placed.

12. The oral implant precision evaluation system according to claim 10, characterized in that, The standard model includes a 3D printed model based on preoperative images of the target object. The 3D printed model includes an initial 3D printed model and a planned 3D printed model. The planned 3D model has reserved implantation positions, and the positions of the implantation positions are determined according to the preoperative planning scheme. The standard preoperative image refers to the image of the initial 3D printed model, and the standard postoperative image refers to the image of the implant or filling material at the implantation site of the planned 3D printed model.

Images