Automated method for registering 3D dental data and computer-readable medium having a program for executing the method

Abstract

An automated method for registering 3D (three-dimensional) dental data, comprising: extracting fiducial points of CT (computed tomography) data; extracting fiducial points of scanned data of a digital impression model; determining an up vector representing the directions of a patient's eyes and nose, and identifying the left and right sides of the fiducial points of the scanned data; extracting a tooth portion of the scanned data; searching for a source point of the scanned data on a spline curve of the CT data to generate a candidate target point group; and determining the candidate target point group having the minimum error with the fiducial points of the CT data as a final candidate.

Description

Technical Field

[0001] An embodiment relates to an automated method for registering 3D (three-dimensional) dental data, and a non-transitory computer-readable medium storing program instructions for an automated method for registering 3D dental data. More specifically, the embodiment relates to an automated method for registering 3D dental data to reduce the time and effort for registering dental CT (computed tomography) images and digital impression models, and a non-transitory computer-readable storage medium storing program instructions for an automated method for registering 3D dental data. Background Art

[0002] In a dental clinic, 3D patient medical image data and 3D digital impression model scan data are used for diagnosis, analysis, and prosthesis fabrication. These two data have different information, and when combined, more effective and various diagnoses, analyses, and fabrications may be achieved. However, since these two data are 3D data defined in different coordinate systems, a registration process for matching these two data is required.

[0003] For the registration of 3D patient medical image data and 3D digital impression model scan data, fiducial points may be set in each of the 3D patient medical image data and the 3D digital impression model scan data.

[0004] Selecting six fiducial points from two 3D data (i.e., dental CT images and digital impression models) may take a lot of time and effort. Additionally, even if fiducial points are found using artificial intelligence techniques, the found fiducial points may not exactly match each other. Further, when there are missing teeth or when the data regions of the two data are different, the initial registration result based only on the fiducial points may not be good. Summary of the Invention

[0005] An embodiment provides an automated method for registering 3D dental data, which can reduce the time and effort for registering dental CT images and digital impression models.

[0006] An embodiment provides a non-transitory computer-readable storage medium storing program instructions for an automated method for registering 3D dental data.

[0007] In an exemplary automated method for registering 3D dental data according to the inventive concept, the method includes: extracting fiducial points of CT (Computed Tomography) data; extracting fiducial points of scan data of a digital impression model; determining an up vector representing the direction of a patient's eyes and nose and identifying the left and right sides of the fiducial points of the scan data; extracting a tooth portion of the scan data; searching for a source point of the scan data on a spline curve of the CT data to generate a candidate target point group; and determining, as a final candidate, the candidate target point group having the minimum error with respect to the fiducial points of the CT data.

[0008] In one embodiment, the fiducial points of the CT data may include three or more fiducial points in the maxilla and three or more fiducial points in the mandible. The fiducial points of the scan data may include three fiducial points.

[0009] In one embodiment, the first and third fiducial points of the scan data may indicate the outermost points of the teeth of the scan data in the lateral direction.

[0010] In one embodiment, when the first fiducial point of the scan data is p 11 , the second fiducial point of the scan data is p 12 , the third fiducial point of the scan data is p 13 , and the average vector obtained by averaging all normal vectors of all points forming the mesh of the scan data is , the left and right sides of the fiducial points of the scan data may be identified using the cross product of and and the average vector .

[0011] In one embodiment, when the scan data represents maxilla data and the discriminant d < 0, the left tooth fiducial point p L representing the outermost left point of the patient's teeth may be p 11 , and the right tooth fiducial point p R representing the outermost right point of the patient's teeth may be p 13 . When the scan data represents the maxilla data and the discriminant d >= 0, the left tooth fiducial point p L may be p 13 , and the right tooth fiducial point p R may be p 11 . The discriminant d may be defined as

[0012] In one embodiment, when the scan data represents mandibular data and the discriminant d < 0, the left tooth landmark point p L can be p 13 , and the right tooth landmark point p R can be p 11 . When the scan data represents the mandibular data and the discriminant d >= 0, the left tooth landmark point p L can be p 11 , and the right tooth landmark point p R can be p 13 .

[0013] In one embodiment, when the upward vector is the left tooth landmark point representing the left outermost point of the patient's teeth is p L , the right tooth landmark point representing the right outermost point of the patient's teeth is p R , the second landmark point of the scan data is p 12 , and the scan data represents the maxillary data,

[0014] In one embodiment, when the upward vector is the left tooth landmark point representing the left outermost point of the patient's teeth is p L , the right tooth landmark point representing the right outermost point of the patient's teeth is p R , the second landmark point of the scan data is p 12 , and the scan data represents the mandibular data,

[0015] In one embodiment, the method may further include determining whether the CT data and the scan data have the same region. When d1 = ||p1 - p3||, d2 = ||p5 - p3||, d3 = ||p 11 - p 12 ||, d4 = ||p 13 - p 12 ||, th is a first threshold for determining that the CT data and the scan data have the same region, p1, p3, and p5 are the landmark points of the CT data, p 11 , p 12 and p 13 are the landmark points of the scan data, and when |(d1 + d2) - (d3 + d4)| < th, the CT data and the scan data are determined to have the same region.

[0016] In one embodiment, the extracting of the tooth portion of the scan data may include: when the scan data represents maxilla data, extracting the highest point among a first landmark point, a second landmark point, and a third landmark point of the scan data in the direction of the upward vector; cutting the scan data into a plane having a normal vector of the upward vector at a point that moves a first distance in the positive direction of the upward vector from the highest point; and cutting the scan data into a plane having a normal vector of the upward vector at a point that moves a second distance in the negative direction of the upward vector from the highest point.

[0017] In one embodiment, the extracting of the tooth portion of the scan data may further include: when the scan data represents mandible data, extracting the lowest point among the first landmark point, the second landmark point, and the third landmark point of the scan data in the direction of the upward vector; cutting the scan data into a plane having a normal vector of the upward vector at a point that moves the first distance in the positive direction of the upward vector from the lowest point; and cutting the scan data into a plane having a normal vector of the upward vector at a point that moves the second distance in the negative direction of the upward vector from the lowest point.

[0018] In one embodiment, the vector from the second landmark point of the scan data to the right tooth landmark point may be defined as and the vector from the second landmark point to the left tooth landmark point may be defined as The extracting of the tooth portion of the scan data may further include: cutting the scan data into a plane having a normal vector of at a point that moves a third distance in the direction of from the right tooth landmark point; and cutting the scan data into a plane having a normal vector of

[0019] In one embodiment, the absolute value of the third distance may be less than the absolute value of the first distance and the absolute value of the second distance.

[0020] In one embodiment, a first vector is defined as the vector rotated -90 degrees, a second vector is defined as the vector rotated +90 degrees, a third vector is defined as the vector rotated +90 degrees, and a fourth vector is defined as the vector A vector rotated -90 degrees. The tooth portion that extracts the scan data may further include: cutting the scan data into a plane having a normal vector of the first vector at a point that moves the third distance in the direction of and moves the fourth distance in the direction of the first vector; cutting the scan data into a plane having a normal vector of the second vector at a point that moves the third distance in the direction of and moves the fourth distance in the direction of the second vector; cutting the scan data into a plane having a normal vector of the third vector at a point that moves the third distance in the direction of from the left tooth landmark point and moves the fourth distance in the direction of the third vector; and cutting the scan data into a plane having a normal vector of the fourth vector at a point that moves the third distance in the direction of from the left tooth landmark point and moves the fourth distance in the direction of the fourth vector.

[0021] In one embodiment, the absolute value of the fourth distance may be greater than the absolute value of the first distance, the absolute value of the second distance, and the absolute value of the third distance.

[0022] In one embodiment, the tooth portion that extracts the scan data may further include: cutting the scan data into a plane having a normal vector of the vector which is the sum of and at a point that moves the fifth distance in the direction of from the second landmark point; and cutting the scan data into a plane having a normal vector of the vector - at a point that moves the fifth distance in the direction of - from the second landmark point.

[0023] In one embodiment, searching for the source point of the scan data on the spline curve of the CT data may include calculating the spline curve C(u) based on multiple landmark points of the maxilla of the CT data or multiple landmark points of the mandible of the CT data.

[0024] In one embodiment, the source point may include three points: the left tooth landmark point, the second landmark point, and the right tooth landmark point.

[0025] In one embodiment, p L may be the left tooth landmark point, p 12 may be the second landmark point, and p RIt can be the right tooth landmark point. The first point C(u1) of the target point can be searched for on C(u) while increasing the parameter u by a first value. The second point C(u2) of the target point can be searched for on C(u) while increasing the parameter u by a second value. The second point C(u2) of the target point can be the point that minimizes d11 = ||C(u1) - C(u2)|| - ||p L - p 12 ||. The third point C(u3) of the target point can be searched for on C(u) while increasing the parameter u by a third value. The third point C(u3) of the target point can be the point that minimizes d12 = ||C(u2) - C(u3)|| - ||p 12 - p R ||. When all of d11, d12, and d13 = ||C(u3) - C(u1)|| - ||p R - p L || are less than a second threshold, the target points C(u1), C(u2), and C(u3) can be selected as the candidate target point group. The target point group can include the three points C(u1), C(u2), and C(u3).

[0026] In one embodiment, determining the candidate target point group having the minimum error with the landmark point of the CT data as the final candidate can include: transforming the candidate target point group to the domain of the CT data using a transformation matrix; and calculating a transformation error as an average of distances between the transformed candidate target point group and the landmark point of the CT data.

[0027] In an example method of an automated method for registering 3D dental data according to the inventive concept, the method includes: extracting landmark points of CT (Computed Tomography) data; extracting landmark points of scan data of a digital impression model; determining an upward vector representing the directions of a patient's eyes and nose and identifying the left and right sides of the landmark points of the scan data; determining whether the CT data and the scan data have the same region; extracting a tooth portion of the scan data; when the CT data and the scan data have different regions, searching for a source point of the scan data on a spline curve of the CT data to generate a candidate target point group; and recommending, as a final candidate, a candidate target point group having the minimum error in registration of the CT data and the scan data among a plurality of candidate target point groups.

[0028] In an example non - transitory computer - readable storage medium having at least one program command stored thereon, the program command is executable by at least one hardware processor to: extract fiducial points of CT data; extract fiducial points of scan data of a digital impression model; determine an up - vector representing the directions of a patient's eyes and nose and identify the left and right sides of the fiducial points of the scan data; extract the tooth portion of the scan data; search for a source point of the scan data on a spline curve of the CT data to generate a group of candidate target points; and determine the group of candidate target points having the minimum error with the fiducial points of the CT data as the final candidate.

[0029] In an example non - transitory computer - readable storage medium having at least one program command stored thereon, the program command is executable by at least one hardware processor to: extract fiducial points of CT (Computed Tomography) data; extract fiducial points of scan data of a digital impression model; determine an up - vector representing the directions of a patient's eyes and nose and identify the left and right sides of the fiducial points of the scan data; determine whether the CT data and the scan data have the same region; extract the tooth portion of the scan data; when the CT data and the scan data have different regions, search for a source point of the scan data on a spline curve of the CT data to generate a group of candidate target points; and recommend the group of candidate target points having the minimum registration error between the CT data and the scan data among multiple groups of candidate target points as the final candidate.

[0030] According to the automated method for registering 3D dental data, good initial registration results can be obtained even without user input and when the data regions are different. Thus, fine - registration results can be quickly obtained without user input.

[0031] According to the automated method for registering 3D dental data, the time and effort for registering patient medical image data (CT or CBCT) and scan data of digital impression models, which are frequently performed in dental clinics and dental laboratories for diagnosis, analysis, and prosthesis fabrication, can be significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and other features and advantages of the inventive concept will become more apparent by referring to the detailed embodiments of the inventive concept described in detail with reference to the accompanying drawings, in which:

[0033] Figure 1 is a flowchart showing an automated method for registering dental CT (Computed Tomography) images and digital impression models according to an embodiment of the inventive concept;

[0034] Figure 2 shows scan data of dental CT images and digital impression models;

[0035] Figure 3 show the fiducial points of the dental CT image;

[0036] Figure 4 show the fiducial points of the scan data of the digital impression model;

[0037] Figures 5 to 8 show the method of determining the up vector and the method of identifying the left and right sides of the fiducial points of the scan data;

[0038] Figure 9 show the up vector when the scan data represents maxilla data;

[0039] Figure 10 show the up vector when the scan data represents mandible data;

[0040] Figure 11 and Figure 12 show Figure 1 the steps of determining whether the region of the CT data and the region of the scan data match;

[0041] Figure 13 and Figure 14 show Figure 1 the steps of extracting the tooth part of the scan data;

[0042] Figure 15 show the tooth part extracted by Figure 1 the steps of extracting the tooth part of the scan data;

[0043] Figure 16 show the tooth part extracted by Figure 1 the steps of extracting the tooth part of the scan data;

[0044] Figure 17 show Figure 1 the steps of searching for the source points of the scan data on the CT spline curve;

[0045] Figures 18A to 18C show Figure 1 the result of the steps of rough registration; and

[0046] Figures 19A to 19C show Figure 1 the result of the steps of fine registration. DETAILED DESCRIPTION

[0047] The present inventive concept will now be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the present invention are shown. However, the present inventive concept may be embodied in many different forms and should not be construed as limited to the exemplary embodiments set forth herein.

[0048] Rather, these exemplary embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Like reference numerals always refer to like elements.

[0049] It should be understood that although the terms first, second, third, etc. may be used herein to describe various elements, components, regions, layers and / or sections, these elements, components, regions, layers and / or sections should not be limited by these terms. These terms are only used to distinguish one element, component, region, layer or section from another region, layer or section. Thus, without departing from the teachings of the present invention, the first element, first component, first region, first layer or first section discussed below may be referred to as a second element, second component, second region, second layer or second section.

[0050] The terms used herein are for the purpose of describing particular exemplary embodiments only and are not intended to limit the present invention. As used herein, the singular forms "a", "an" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that when used in this specification, the terms "comprise" and / or "comprising" specify the presence of the stated features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0051] All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs, unless otherwise defined. It will also be understood that terms, such as those defined in a commonly used dictionary, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0052] Unless otherwise specified herein or clearly contradicted by context, all methods described herein can be performed in a suitable order. Unless otherwise specified, the use of any and all example or exemplary language (e.g., "such as") is only intended to better illustrate the present invention and does not limit the scope of the present invention. Any language in the specification should not be construed as indicating any non-claimed element as essential for the practice of the inventive concepts as used herein.

[0053] Hereinafter, the inventive concept will be explained in detail with reference to the drawings.

[0054] Figure 1 is a flowchart showing an automated method for registering dental CT (Computed Tomography) images and digital impression models according to an embodiment of the inventive concept.Figure 2 Shows the dental CT image and the scan data of the digital impression model. Figure 3 Shows the fiducial points of the dental CT image. Figure 4 Shows the fiducial points of the scan data of the digital impression model.

[0055] Reference Figures 1 to 4 , for the registration of the dental CT image and the digital impression image, fiducial points of the dental CT image can be extracted (step S100) and fiducial points of the digital impression image can be extracted (step S200).

[0056] For example, the dental CT image can be a cone beam CT (CBCT) image. The dental CT image can include teeth, bone, and nerve canals. For example, the scan data of the digital impression model can be obtained by scanning the inside of the patient's mouth with a scanner. For example, the scan data can be obtained by scanning a plaster model of the inside of the patient's mouth with a scanner.

[0057] Figure 2 The left image of can be the scan data of the digital impression model. Figure 2 The right image of can be the dental CT image. In this embodiment, the digital impression model can correspond to one of the patient's maxilla and mandible. In this embodiment, the dental CT image can include information of both the patient's maxilla and mandible.

[0058] For example, in Figure 3 , the fiducial points of the dental CT image can indicate the specific positions of the teeth. For example, the fiducial points of the dental CT image can include three or more fiducial points in the maxilla and three or more fiducial points in the mandible. The fiducial points of the dental CT image can include five fiducial points p1, p2, p3, p4, and p5 in the maxilla and five fiducial points p6, p7, p8, p9, and p10 in the mandible. For example, the first fiducial point p1 of the maxilla and the fifth fiducial point p5 of the maxilla can indicate the outermost points of the teeth of the maxilla in the transverse direction. The third fiducial point p3 of the maxilla can indicate between the two maxillary central incisors. The second fiducial point p2 of the maxilla can be set between the first fiducial point p1 and the third fiducial point p3. The fourth fiducial point p4 of the maxilla can be set between the third fiducial point p3 and the fifth fiducial point p5. For example, the sixth fiducial point p6 of the mandible and the tenth fiducial point p10 of the mandible can indicate the outermost points of the teeth of the mandible in the transverse direction. The eighth fiducial point p8 of the mandible can indicate between the two mandibular central incisors. The seventh fiducial point p7 of the mandible can be set between the sixth fiducial point p6 and the eighth fiducial point p8. The ninth fiducial point p9 of the mandible can be set between the eighth fiducial point p8 and the tenth fiducial point p10.

[0059] For example, in Figure 4Among them, the landmark points of the scan data can indicate the specific position of the teeth. The landmark points of the scan data can include three landmark points p 11 , p 12 and p 13 . In this article, the scan data can represent the maxilla or the mandible of the patient. For example, the first landmark point p 11 and the third landmark point p 13 of the scan data can indicate the outermost points of the teeth in the scan data in the lateral direction. The second landmark point p 12 of the scan data can indicate between the two central incisors.

[0060] In this embodiment, the artificial intelligence deep learning method can be used to automatically extract the landmark points (e.g., p1 to p10) of the CT image. Additionally, the artificial intelligence deep learning method can be used to automatically extract the landmark points (e.g., p 11 to p 13 ) of the scan data. Whether the scan data represents the maxilla or the mandible can be determined by the user's input, or can be automatically determined using additional information of the scan data.

[0061] Figures 5 to 8 shows a method for determining the upward vector and a method for identifying the left and right sides of the landmark points of the scan data.

[0062] Refer to Figures 5 to 8 to explain a method for determining the upward vector representing the upward direction (the direction of the patient's eyes and nose) of the scan data and a method for identifying the left and right sides of the landmark points p 11 to p 13 of the scan data (step S300).

[0063] For example, in Figure 5 , the teeth protrude downward in the scan data, the first landmark point p 11 is set in the left part of the scan data, and the third landmark point p 13 is set in the right part of the scan data. Even if Figure 5 the teeth in the scan data protrude downward, Figure 5 the scan data does not necessarily mean maxilla data. As explained above, whether the scan data represents the maxilla or the mandible can be determined by the user's input or by using additional information of the scan data.

[0064] The normal vector can be obtained from all points (all points on the surface of the scan data) forming the grid of the scan data. The unit vector of the average vector is obtained by averaging all the normal vectors from all points Figure 5 . The average vector obtained by averaging the normal vectors from all points may point downward in Figure 5 . The unit vector of the average vector may have a unit length in the downward direction.

[0065] The discriminant for identifying the left and right sides of the landmark points of the scan data is shown in Equation 1 below.

[0066] [Equation 1]

[0067]

[0068] When the scan data represents maxilla data and the discriminant d < 0, the left tooth landmark point p representing the leftmost outer point of the patient's teeth L may be p 11 , and the right tooth landmark point p representing the rightmost outer point of the patient's teeth R may be p 13 . Conversely, when the scan data represents maxilla data and the discriminant d >= 0, the left tooth landmark point p L may be p 13 , and the right tooth landmark point p R may be p 11 .

[0069] When the scan data represents mandible data and the discriminant d < 0, the left tooth landmark point p representing the leftmost outer point of the patient's teeth L may be p 13 , and the right tooth landmark point p representing the rightmost outer point of the patient's teeth R may be p 11 . Conversely, when the scan data represents mandible data and the discriminant d >= 0, the left tooth landmark point p L may be p 11 , and the right tooth landmark point p R may be p 13 .

[0070] In Figure 5 , since has a downward direction, when the upward direction should be the positive direction, and 's cross product may have a downward direction (negative value). Thus, according to Equation 1, d is equal to or greater than zero (d >= 0), such that when the scan data in Figure 5 represents maxilla data, the left landmark point p L = p 13 and the right tooth landmark point p R = p11 . In contrast, according to Equation 1, d is equal to or greater than zero (d >= 0), such that when Figure 5 the scanned data in L represents mandibular data, the left fiducial point p 11 = p R and the right tooth fiducial point p 13 .

[0071] For example, in Figure 6 the teeth protrude downward in the scanned data, the third fiducial point p 13 is set in the left part of the scanned data, and the first fiducial point p 11 is set in the right part of the scanned data.

[0072] In Figure 6 , since has a downward direction, when the upward direction should be the positive direction, and the cross product of can have an upward direction (positive value). Thus, according to Equation 1, d is less than zero (d < 0), such that when Figure 6 the scanned data in L represents maxillary data, the left fiducial point p 11 = p R and the right tooth fiducial point p 13 = p L . In contrast, according to Equation 1, d is less than zero (d < 0), such that when 13 the scanned data in R represents mandibular data, the left fiducial point p 11 = p 13 .

[0073] For example, in Figure 7 the teeth protrude upward in the scanned data, the third fiducial point p 11 is set in the left part of the scanned data, and the first fiducial point p L is set in the right part of the scanned data.

[0074] In Figure 7 , since has an upward direction, when the upward direction should be the positive direction, and the cross product of can have an upward direction (positive value). Thus, according to Equation 1, d is equal to or greater than zero (d >= 0), such that when Figure 7 the scanned data inL = p 11 and the right tooth landmark point p R = p 13 . In contrast, according to Equation 1, d is equal to or greater than zero (d >= 0), such that when Figure 7 the scan data in represents maxilla data, the left landmark point p L = p 13 and the right tooth landmark point p R = p 11 .

[0075] For example, in Figure 8 the teeth protrude upward in the scan data, the first landmark point p 11 is set in the left part of the scan data, and the third landmark point p 13 is set in the right part of the scan data.

[0076] In Figure 8 since has an upward direction, when the upward direction should be the positive direction, and the cross product of can have a downward direction (negative value). Therefore, according to Equation 1, d is less than zero (d < 0), such that when Figure 8 the scan data in represents mandible data, the left landmark point p L = p 13 and the right tooth landmark point p R = p 11 . In contrast, according to Equation 1, d is less than zero (d < 0), such that when Figure 8 the scan data in represents maxilla data, the left landmark point p L = p 11 and the right tooth landmark point p R = p 13 .

[0077] When the scan data represents maxilla data, the upward vector representing the direction of the patient's eyes and nose can be expressed as the following Equation 2.

[0078] [Equation 2]

[0079]

[0080] For example, when Figure 5 the scan data of represents maxilla data, according to the discriminant, the left landmark point p L = p 13 and the right tooth landmark point p R = p 11According to Equation 2, as p R -p 12 and p L -p 12 The cross product of can have an upward direction (positive value).

[0081] When the scan data represents mandibular data, the upward vector indicating the directions of the patient's eyes and nose can be represented by the following Equation 3.

[0082] [Equation 3]

[0083]

[0084] For example, when Figure 7 the scan data represents mandibular data, according to the discriminant, the left landmark point p L = p 11 and the right tooth landmark point p R = p 13 According to Equation 3, as p L -p 12 and p R -p 12 The cross product of can have an upward direction (positive value).

[0085] Figure 9 Shows the upward vector when the scan data represents maxillary data. Figure 10 Shows the upward vector when the scan data represents mandibular data.

[0086] As Figure 9 shown, when the scan data represents maxillary data, the direction of the upward vector can be opposite to the protruding direction of the teeth.

[0087] As Figure 10 shown, when the scan data represents mandibular data, the direction of the upward vector can be the same as the protruding direction of the teeth.

[0088] Figure 11 And Figure 12 shows Figure 1 The steps for determining whether the region of the CT data and the region of the scan data match.

[0089] Refer to Figures 1 to 12, in the registration of CT data and scan data, case 1 and case 2 can be distinguished (step S400). Case 1 means that the CT data and the scan data have the same region. Case 2 means that the CT data and the scan data have different regions. When the following Equation 4 is satisfied, it can be determined that the CT data and the scan data have the same region.

[0090] [Equation 4]

[0091] |(d1 + d2)-(d3 + d4)| < th

[0092] In this article, d1 = ||p1 - p3||, d2 = ||p5 - p3||, d3 = ||p 11 -p 12 ||, d4 = ||p 13 -p 12 ||, and th is the first threshold value used to determine that the CT data and the scan data have the same region. For example, the first threshold value th can be 5 mm.

[0093] Figure 13 and Figure 14 shows Figure 1 the steps of extracting the tooth part of the scan data. Figure 15 shows Figure 1 the tooth part extracted by the steps of extracting the tooth part of the scan data through

[0094] Referring to Figures 1 to 15 , in order to perform registration based on the tooth part that is the common region between the two data, only the tooth part can be cut out from the scan data (step S500). When the scan data represents maxilla data, the highest point among the first landmark point p , the second landmark point p 11 , and the third landmark point p 12 is extracted in the direction of the upward vector 13 . When the scan data represents mandible data, the lowest point among the first landmark point p , the second landmark point p 11 , and the third landmark point p 12 , and the third landmark point p 13 is extracted in the direction of the upward vector

[0095] When the scan data represents maxilla data, the scan data can be cut into an infinite plane having a normal vector at a point that moves a first distance (+a) in the positive direction of the upward vector from the highest point of the scan data, and the scan data can be cut into an infinite plane having a normal vector at a point that moves a second distance (-b) in the negative direction of the upward vector from the highest point of the scan data, and the scan data can be cut into an infinite plane having a normal vector at a point that moves a second distance (-b) in the negative direction of the upward vector from the highest point of the scan data, and the scan data can be cut into an infinite plane having a normal vector at a point that moves a second distance (-b) in the negative direction of the upward vector infinite plane of the normal vector ( Figure 13 ). For example, the first distance (+a) can be 6 mm. For example, the second distance (-b) can be -6 mm. For example, the absolute value of the first distance (+a) can be equal to the absolute value of the second distance (-b). Alternatively, the absolute value of the first distance (+a) can be different from the absolute value of the second distance (-b).

[0096] When the scan data represents mandibular data, the scan data can be cut into an infinite plane CP1 having a normal vector at a point that moves the first distance (+a) in the positive direction of the upward vector from the lowest point of the scan data , and the scan data can be cut into an infinite plane CP2 having a normal vector at a point that moves the second distance (-b) in the negative direction of the upward vector from the lowest point of the scan data ( Figure 13 ).

[0097] As Figure 14 shown, the vector from the second landmark point p 12 to the right tooth landmark point p R can be defined as and the vector from the second landmark point p 12 to the left tooth landmark point p L can be defined as The scan data can be cut into an infinite plane CP3 having a R normal vector at a point that moves a third distance in the direction of from the right tooth landmark point p. The scan data can be cut into an infinite plane CP4 having a normal vector at a point that moves a third distance in the direction of L from the left tooth landmark point p . For example, the absolute value of the third distance can be less than the absolute values of the first distance and the second distance. In this article, the third distance can be 1 mm.

[0098] The first vector can be defined as the vector obtained by rotating the vector by -90 degrees. The scan data can be cut into an infinite plane CP5 having a normal vector of the first vector at a point that moves a third distance in the direction of from the right tooth landmark point p R and moves a fourth distance in the direction of the first vector. The second vector can be defined as the vector obtained by rotating the vector by +90 degrees. The scan data can be cut into an infinite plane CP5 having a normal vector of the first vector at a point that moves a third distance in the direction of from the right tooth landmark point p R and...An infinite plane CP6 having a normal vector of the second vector at a point that moves a third distance in the direction of the first vector and a fourth distance in the direction of the second vector. For example, the absolute value of the fourth distance may be greater than the absolute values of the first distance, the second distance, and the third distance. In this context, the fourth distance may be 10 mm.

[0099] The third vector may be defined as a vector rotated +90 degrees from the vector The scan data may be cut into an infinite plane CP7 having a normal vector of the third vector at a point that moves a third distance in the direction of L along and a fourth distance in the direction of the third vector. The fourth vector may be defined as a vector rotated -90 degrees from the vector The scan data may be cut into an infinite plane CP8 having a normal vector of the fourth vector at a point that moves a third distance in the direction of L along and a fourth distance in the direction of the fourth vector.

[0100] In addition, the scan data may be cut into an infinite plane CP10 having a normal vector of the vector 12 along which is the sum of and at a point that moves a fifth distance in the direction of The scan data may be cut into an infinite plane CP9 having a normal vector of the vector - 12 along - at a point that moves a fifth distance in the direction of For example, the absolute value of the fifth distance may be greater than the absolute value of the third distance and less than the absolute value of the fourth distance. In this context, the fifth distance may be 6 mm.

[0101] Figure 15 Denotes a tooth portion obtained by cutting the scan data using the cutting planes CP1 to CP10.

[0102] The first fiducial points p1 to p5 of the maxilla of the CT data may be used as control points to calculate the parametric spline curve C(u). In this context, u satisfies 0 <= u <= 1. At the far left of the patient, u = 0, and at the far right of the patient, u = 1.

[0103] The parametric spline curve C(u) may refer to a spline curve of an arc connecting the five fiducial points p1 to p5 of the maxilla of the CT data. Alternatively, the five fiducial points p6 to p10 of the mandible of the CT data may be used as control points to calculate the parametric spline curve C(u).

[0104] The source points of the scanned data can be searched on the CT spline curve C(u) to generate target points (step S600).

[0105] The source points of the scanned data can include the left tooth landmark point p L , the second landmark point p 12 and the right tooth landmark point p R These three points.

[0106] While increasing the parameter u by a first value, the first point of the target point can be searched on C(u). The first point of the target point can be represented as C(u1). The first value can be 0.05.

[0107] While increasing the parameter u by a second value, the second point of the target point can be searched on C(u). The second point of the target point can be represented as C(u2). The second point C(u2) of the target point can be the point that minimizes d11 = ||C(u1) - C(u2)|| - ||p L - p 12 ||. Herein, u > u1 and the second value can be 0.001.

[0108] While increasing the parameter u by a third value, the third point of the target point can be searched on C(u). The third point of the target point can be represented as C(u3). The third point C(u3) of the target point can be the point that minimizes d12 = ||C(u2) - C(u3)|| - ||p 12 - p R ||. Herein, u > u2 and the third value can be 0.001.

[0109] When all of d11, d12, and d13 = ||C(u3) - C(u1)|| - ||p R - p L || are less than a second threshold, the target points C(u1), C(u2), and C(u3) can be selected as candidates. The second threshold can be 8 mm.

[0110] Figure 16 Shows the tooth portion extracted by Figure 1 the steps of extracting the tooth portion of the scanned data.

[0111] Refer to Figures 1 to 16 , in this embodiment, case 2 where the CT data and the scanned data have different regions is shown. For example, the CT data can include the entire region of the patient's teeth, while the scanned data can include only a part of the patient's teeth. Although, for example, in this embodiment, the CT data includes the entire region of the patient's teeth and the scanned data includes only a part of the patient's teeth, the inventive concept is not limited thereto.

[0112] When Equation 4 is satisfied, the CT data and the scan data can be identified as Case 1 where the CT data and the scan data have the same region. When Equation 4 is not satisfied, the CT data and the scan data can be identified as Case 2 where the CT data and the scan data have different regions. In this embodiment, Equation 4 may not be satisfied, such that the CT data and the scan data may have different regions.

[0113] As Figure 16 shown, the scan data may include a part of the patient's teeth.

[0114] Figure 17 Illustrated is Figure 1 the step of searching for the source point of the scan data on the search CT spline curve.

[0115] Referring Figures 1 to 17 , as explained in the reference Figure 15 , in Figure 16 , the first fiducial point p1 to the fifth fiducial point p5 of the maxilla of the CT data can be used as control points to calculate the parametric spline curve C(u).

[0116] For Figure 16 the scan data, the source point of the scan data can be searched on the CT spline curve C(u) to generate a target point (step S600).

[0117] The source point of the scan data may include the left tooth fiducial point p L , the second fiducial point p 12 and the right tooth fiducial point p R these three points.

[0118] While increasing the parameter u by a first value, the first point of the target point can be searched on C(u). The first point of the target point can be expressed as C(u1). The first value can be 0.05.

[0119] While increasing the parameter u by a second value, the second point of the target point can be searched on C(u). The second point of the target point can be expressed as C(u2). The second point C(u2) of the target point can be the point that minimizes d11 = ||C(u1) - C(u2)|| - ||p L - p 12 ||. Here, u > u1 and the second value can be 0.001.

[0120] While increasing the parameter u by a third value, the third point of the target point can be searched on C(u). The third point of the target point can be expressed as C(u3). The third point C(u3) of the target point can be the point that minimizes d12 = ||C(u2) - C(u3)|| - ||p 12 - p R||Minimized points. In this article, u > u2 and the third value can be 0.001.

[0121] When all of d11, d12, and d13 = ||C(u3) - C(u1)|| - ||p R -p L ||are all less than the second threshold, the target points C(u1), C(u2), and C(u3) can be selected as candidates. The second threshold can be 8 mm.

[0122] Figures 18A to 18C Shows Figure 1 the result of the steps of rough registration.

[0123] Reference Figures 1 to 18C , the candidate target point group can include six points, and multiple candidate target point groups can be generated in step S500.

[0124] The transformation matrix M can be calculated by the flag transformation for multiple candidate target point groups. The transformation error can be calculated as the average of the distances between the transformed flag points pi' = Mpi (i = L, 12, R) of the scan data and the flag points pk (k = 1, 3, 5 or k = 6, 8, 10) of the CT data. The transformation matrix M can move the flag points of the scan data to the domain of the CT data.

[0125] In step S400, in case 1 where the CT data and the scan data have the same region, the candidate target point group with the minimum transformation error can be determined as the final candidate.

[0126] The step of moving the flag points of the scan data to the domain of the CT data and the step of determining the final candidate with the minimum transformation error can be called rough registration (step S700).

[0127] In step S400, in case 2 where the CT data and the scan data have different regions, the candidate target point group with the minimum transformation error may not be determined as the final candidate because, in case 2, the left tooth flag point p L , the second flag point p 12 and the right tooth flag point p R do not correspond to the first flag p1, the third flag p3, and the fifth flag p5 of the CT data.

[0128] Instead, in case 2, the final candidate can be determined in the steps of fine registration (step S800) explained below.

[0129] In Figures 18A to 18C , the background image representing the shape of the teeth can be the CT data. In Figures 18A to 18C , the solid line can be the contour of the scan data generated by registering with the CT data.Figure 18A An axial view after rough registration of CT data and scan data is shown. Figure 18B A sagittal view after rough registration of CT data and scan data is shown. Figure 18C A coronal view after rough registration of CT data and scan data is shown.

[0130] Figures 19A to 19C Shows Figure 1 the result of the steps of fine registration.

[0131] Refer to Figures 1 to 19C , after rough registration (step S700), fine registration (step S800) for further matching the tooth region of the CT data with the tooth region of the scan data can be performed. In the fine registration, the erupted part of the scan data can be used as the source data, and the CT image of the patient can be used as the target data.

[0132] In Figures 19A to 19C , the background image representing the shape of the teeth can be CT data. In Figures 19A to 19C , the solid line can be the contour of the scan data generated by registering with the CT data. Figure 19A An axial view after fine registration of CT data and scan data is shown. Figure 18B A sagittal view after fine registration of CT data and scan data is shown. Figure 18C A coronal view after fine registration of CT data and scan data is shown.

[0133] As Figures 19A to 19C shown, compared with Figures 18A to 18C , the tooth part of the CT data and the tooth part of the scan data can be more precisely matched.

[0134] In case 2 where the CT data and the scan data have different regions, fine registration can be performed for multiple candidate target point groups. The final candidate among the multiple candidate target point groups can be recommended to the user by arranging the results of the fine registration in the order of the minimum error.

[0135] The user can determine one of the candidate target point groups as the final candidate based on the error obtained through the fine registration.

[0136] According to this embodiment, even without user input and with different data regions, a good initial registration result can be obtained. Therefore, the result of the fine registration can be quickly obtained without user input.

[0137] According to an automated method for registering 3D dental data, the time and effort for registering patient medical image data (CT or CBCT) and digital impression model scan data, which are frequently performed for diagnosis, analysis, and prosthesis fabrication in dental clinics and dental laboratories, can be significantly reduced.

[0138] According to an embodiment of the inventive concept, a non-transitory computer-readable storage medium can be provided, on which program instructions for an automated method for registering 3D dental data are stored. The above method can be written as a program to be executed on a computing device (such as a computer). The method can be implemented in a general digital computer that operates a program using a computer-readable medium. Additionally, the structure of the data used in the above method can be written on the computer-readable medium in various ways. The computer-readable medium can include program instructions, data files, and data structures, either alone or in combination. The program instructions written on the medium can be specifically designed and configured for the inventive concept, or can be well-known to those skilled in the field of computer software. For example, the computer-readable medium can include magnetic media (such as hard disks, floppy disks, and magnetic tapes), optical media (such as CD-ROMs and DVDs), magneto-optical media (such as optical disks), and hardware devices (such as ROM, RAM, and flash memory) specifically configured to store and execute program instructions. For example, the program instructions can include machine language code generated by a compiler, and high-level language code that can be executed by a computer using an interpreter, etc. The hardware device can be configured to operate as one or more software modules to perform the operations of the inventive concept.

[0139] Additionally, the above method for registering 3D dental data can be implemented in the form of a computer-executable computer program or an application program stored in a storage method.

[0140] The inventive concept relates to an automated method for registering 3D dental data and a non-transitory computer-readable storage medium on which program instructions for an automated method for registering 3D dental data are stored, which can reduce the time and effort for registering dental CT data and digital impression models.

[0141] The foregoing is a description of the inventive concept and should not be construed as a limitation thereof. Although several embodiments of the inventive concept have been described, those skilled in the art will readily appreciate that many modifications may be made to the embodiments without departing from the novel teachings and advantages of the inventive concept in essence. Accordingly, all such modifications are intended to be included within the scope of the inventive concept as defined by the claims. In the claims, the means-plus-function clauses are intended to cover the structures that perform the recited functions herein, and include not only structural equivalents but also equivalent structures. Therefore, it should be understood that the foregoing is a description of the inventive concept and should not be construed as limited to the specific embodiments disclosed, and that modifications to the disclosed embodiments as well as other embodiments are intended to be included within the scope of the appended claims. The inventive concept is defined by the following claims, and equivalents of the claims are included therein.

Claims

1. An automated method for registering 3D (three-dimensional) dental data, the method comprising: Extracting fiducial points of CT (computed tomography) data; Extracting fiducial points of the scanned data of the digital impression model; Determining an upward vector representing the directions of the patient's eyes and nose and identifying a left tooth fiducial point representing the leftmost outermost point of the patient's teeth and a right tooth fiducial point representing the rightmost outermost point of the patient's teeth in the scanned data; Extracting the tooth portion of the scanned data; Searching for a source point of the scanned data on a spline curve of the CT data to generate a group of candidate target points; And Determining the group of candidate target points having the minimum error with the fiducial points of the CT data as the final candidates; Wherein extracting the tooth portion of the scanned data comprises: When the scanned data represents maxilla data, extracting the highest point among the first fiducial point, the second fiducial point, and the third fiducial point of the scanned data in the direction of the upward vector, Cutting the scanned data into a plane having a normal vector of the upward vector at a point that moves a first distance in the positive direction of the upward vector from the highest point; and Cutting the scanned data into a plane having a normal vector of the upward vector at a point that moves a second distance in the negative direction of the upward vector from the highest point; When the scanned data represents mandible data, extracting the lowest point among the first fiducial point, the second fiducial point, and the third fiducial point of the scanned data in the direction of the upward vector, Cutting the scanned data into a plane having a normal vector of the upward vector at a point that moves the first distance in the positive direction of the upward vector from the lowest point; and Cutting the scanned data into a plane having a normal vector of the upward vector at a point that moves the second distance in the negative direction of the upward vector from the lowest point; wherein the vector from the second landmark point of the scan data to the right tooth landmark point is defined as and the vector from the second landmark point to the left tooth landmark point is defined as Cut the scan data into a plane having a normal vector at a point that moves a third distance in the direction of from the right tooth landmark point; and and; Cut the scan data into a plane having a normal vector at a point that moves the third distance in the direction of from the left tooth landmark point .

2. The method according to claim 1, wherein the fiducial points of the CT data include three or more fiducial points in the maxilla and three or more fiducial points in the mandible, and wherein the fiducial points of the scanned data include three fiducial points.

3. The method according to claim 2, wherein the first fiducial point and the third fiducial point of the scanned data indicate the outermost points of the teeth in the scanned data in the lateral direction.

4. The method according to claim 1, wherein when the first landmark point of the scan data is p 11 , the second landmark point of the scan data is p 12 , the third landmark point of the scan data is p 13 , and the average vector obtained by averaging all the normal vectors of all the points from the grid forming the scan data is , use and the cross product of and the average vector to identify the left tooth landmark point and the right tooth landmark point of the scan data.

5. The method according to claim 4, wherein when the scan data represents maxilla data and the discriminant d < 0, the left tooth landmark point p of the leftmost outer point of the teeth of the patient is L p 11 , and the right tooth landmark point p of the rightmost outer point of the teeth of the patient is R p 13 , When the scan data represents the maxilla data and the discriminant d >= 0, the left tooth landmark point p L is p 13 , and the right tooth landmark point p R is p 11 , and wherein the discriminant d is defined as 6. The method according to claim 5, wherein when the scanned data represents mandibular bone data and the discriminant d < 0, the left tooth landmark point p L is p 13 , and the right tooth landmark point p R is p 11 , and Wherein when the scan data represents the mandibular bone data and the discriminant d >= 0, the left tooth landmark point p L is p 11 , and the right tooth landmark point p R is p 13 .

7. The method according to claim 1, wherein when the upward vector is the left tooth landmark point representing the left outermost point of the teeth of the patient is p L , the right tooth landmark point representing the right outermost point of the teeth of the patient is p R , the second landmark point of the scan data is p 12 , and when the scan data represents the maxilla data, 8. The method according to claim 1, wherein when the upward vector is the left tooth landmark point representing the leftmost outer point of the teeth of the patient is p L , the right tooth landmark point representing the rightmost outer point of the teeth of the patient is p R , the second landmark point of the scan data is p 12 , and when the scan data represents mandibular data, 9. The method according to claim 1, the method further comprising determining whether the CT data and the scanned data have the same region, where when d1 = ||p1 - p3||, d2 = ||p5 - p3||, d3 = ||p 11 - p 12 ||, d4 = ||p 13 - p 12 ||, th is a first threshold value for determining that the CT data and the scan data correspond to the same data region of the patient, p1, p3 and p5 are the fiducial points of the CT data, p 11 , p 12 and p 13 are the fiducial points of the scan data, and when |(d1 + d2) - (d3 + d4)| < th, the CT data and the scan data are determined to correspond to the same data region of the patient.

10. The method according to claim 1, wherein the absolute value of the third distance is less than the absolute value of the first distance and the absolute value of the second distance.

11. The method according to claim 1, wherein the first vector is defined as the vector obtained by rotating the vector by -90 degrees, the second vector is defined as the vector obtained by rotating the vector by +90 degrees, the third vector is defined as the vector obtained by rotating the vector by +90 degrees, and the fourth vector is defined as the vector obtained by rotating the vector by -90 degrees. Wherein extracting the tooth portion of the scanned data further comprises: Cut the scan data into a plane having a normal vector of the first vector at a point that moves the third distance in the direction of from the right tooth landmark point and moves the fourth distance in the direction of the first vector; ​ Cut the scan data into a plane having a normal vector of the second vector at a point that moves the third distance in the direction from the right tooth landmark point along and moves the fourth distance in the direction of the second vector; Cut the scan data into a plane having a normal vector of the third vector at a point that moves the third distance in the direction from the left tooth landmark point along and moves the fourth distance in the direction of the third vector; and Cut the scan data into a plane having a normal vector of the fourth vector at a point that moves the third distance in the direction from the left tooth landmark point along and moves the fourth distance in the direction of the fourth vector.

12. The method according to claim 11, wherein the absolute value of the fourth distance is greater than the absolute value of the first distance, the absolute value of the second distance, and the absolute value of the third distance.

13. The method according to claim 1, wherein extracting the tooth portion of the scanned data further comprises: Cut the scan data into a plane having a normal vector that is the sum of at a point that moves a fifth distance in the direction of and as a vector ; and Cut the scan data into a plane having a normal vector of vector - at a point that moves the fifth distance in the direction from the second fiducial point along - . in the direction of the fifth distance.

14. The method according to claim 1, wherein searching for the source point of the scan data on the spline curve of the CT data comprises: Calculate the spline curve C(u) based on multiple landmark points of the maxilla of the CT data or multiple landmark points of the mandible of the CT data.

15. The method according to claim 14, wherein the source points include three points: a left tooth landmark point, the second landmark point, and a right tooth landmark point.

16. The method according to claim 15, wherein p L is the left tooth landmark point, p 12 is the second landmark point, and p R is the right tooth landmark point, Among them, while increasing the parameter u by a first value, the first point C(u1) of the target point is searched on C(u). While increasing the parameter u by a second value, the second point C(u2) of the target point is searched on C(u). The second point C(u2) of the target point is the point that minimizes d11 = ||C(u1) - C(u2)|| - ||p L - p 12 ||. While increasing the parameter u by a third value, the third point C(u3) of the target point is searched on C(u). The third point C(u3) of the target point is the point that minimizes d12 = ||C(u2) - C(u3)|| - ||p 12 - p R ||. where when all of d11, d12, and d13 = ||C(u3) - C(u1)|| - ||p R -p L || are less than the second threshold, the target points C(u1), C(u2), and C(u3) are selected as the candidate target point group, and Wherein the target point group includes three points: C(u1), C(u2), and C(u3).

17. The method according to claim 14, wherein determining the candidate target point group having the minimum error with the landmark points of the CT data as the final candidate includes: Transforming the candidate target point group to the domain of the CT data using a transformation matrix; And Calculating the transformation error as the average of the distances between the transformed candidate target point group and the landmark points of the CT data.

18. An automated method for registering 3D (three-dimensional) dental data, the method comprising: Extracting landmark points of CT (computed tomography) data; Extracting landmark points of scan data of a digital impression model; Determining an upward vector representing the direction of the patient's eyes and nose, and identifying a left tooth landmark point representing the leftmost outer point of the patient's teeth and a right tooth landmark point representing the rightmost outer point of the patient's teeth in the scan data; Determining whether the CT data and the scan data have the same region; Extracting the tooth part of the scan data; When the CT data and the scan data have different regions, searching for source points of the scan data on the spline curve of the CT data to generate a candidate target point group; and Recommending, as the final candidate, the candidate target point group having the minimum error in the registration of the CT data and the scan data among multiple candidate target point groups, Wherein extracting the tooth part of the scan data includes: When the scan data represents maxilla data, extracting the highest point among the first landmark point, the second landmark point, and the third landmark point of the scan data in the direction of the upward vector, Cutting the scan data into a plane having a normal vector of the upward vector at a point that moves a first distance in the positive direction of the upward vector from the highest point; and Cutting the scan data into a plane having a normal vector of the upward vector at a point that moves a second distance in the negative direction of the upward vector from the highest point; When the scan data represents mandible data, extracting the lowest point among the first landmark point, the second landmark point, and the third landmark point of the scan data in the direction of the upward vector, Cutting the scan data into a plane having a normal vector of the upward vector at a point that moves the first distance in the positive direction of the upward vector from the lowest point; and Cutting the scan data into a plane having a normal vector of the upward vector at a point that moves the second distance in the negative direction of the upward vector from the lowest point; The vector from the second landmark point of the scan data to the right tooth landmark is defined as and the vector from the second landmark point to the left tooth landmark is defined as Cut the scan data into a plane having a normal vector at a point that moves a third distance in the direction of from the right tooth landmark point; and and Cut the scan data into a plane having a normal vector at a point that moves the third distance in the direction of from the left tooth landmark point. ​ 19. A non-transitory computer-readable storage medium having stored thereon at least one program including instructions that, when executed by at least one hardware processor, perform the method according to any one of claims 1 and 18.

Images