KR20190090663A - Method of automatically generating a curve corresponding to an dental arch shape on dental ct slice automatically selected from three dimensional dental ct image - Google Patents

Abstract

The present invention relates to a method of automatically generating a curve corresponding to a dental arch shape on a dental CT slice automatically selected from a three dimensional dental CT image. The method includes: a step of setting a threshold contrast value for a voxel representing a tooth based on an externally input three dimensional dental CT image, and generating a point cloud storing coordinates of the corresponding voxels only for a voxel having a contrast value greater than the threshold contrast value; a step of selecting a CT slice that best represents the feature of a dental arch shape from the plurality of CT slices constituting the input three dimensional dental CT image as a reference CT slice, based on the point cloud; and a step of arranging a plurality of main control points on the dental arch shape shown in the reference CT slice and generating the curve along the plurality of main control points. It is possible to improve the efficiency of dental diagnosis.

Description

 {METHOD OF AUTOMATICALLY GENERATING A CURVE CORRESPONDING TO AN DENTAL ARCH SHAPE ON DENTAL CT SLICE AUTOMATICALLY SELECTED FROM THREE DIMENSIONAL DENTAL CT IMAGE}

The present invention relates to a method for automatically generating a curve corresponding to a arch form on a tooth CT slice. More specifically, when reconstructing a panoramic image or a tooth section image from a 3D tooth CT image, a curve is drawn. It is a method for automatically generating a curve along a arch by selecting a CT slice.

In general, the user of the software and apparatus for processing three-dimensional dental CT images has to go through a two-step process for panoramic reconstruction and observing the tooth cross section. First, based on the image displayed on the display, a CT slice for drawing a curve from a three-dimensional CT volume image composed of a plurality of CT slices should be selected as an input interface. Secondly, the reference points must be entered one by one from the corresponding CT slices displayed on the display. Based on these entered points, the diagnostic software curves them to create a 'curve'. FIG. 1A is a curve created by a point directly input by a user along the dental arch of the patient, FIG. 1B is a longitudinal section of a tooth orthogonal to the curve of FIG. 1A, and FIG. 1C is a reconstructed vertical section along a curve Represents a panoramic image.

Currently, curve generation for panoramic image and cross-sectional reconstruction to be useful for diagnosis depends entirely on the user's skill, and for the inexperienced user, it takes a lot of time and effort to express the ideal image for diagnosis. have.

An object of the present invention is to improve the efficiency of the overall dental diagnosis by automatically selecting a reference CT slice for generating a curve from a three-dimensional dental CT image composed of several CT slices.

The present invention aims to reduce the time and effort to obtain accurate results in using diagnostic software and devices by automatically generating curves in selected CT slices.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention, which are not mentioned above, can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. It will also be readily apparent that the objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.

The present invention for achieving the above object, in a method for automatically generating a curve corresponding to the arch shape on the tooth CT slice automatically selected from the three-dimensional CT image, based on the three-dimensional teeth CT image input from the outside Setting a threshold contrast value for a voxel indicating a value, and generating a point cloud storing coordinates of corresponding voxels only for a voxel having a contrast value greater than the threshold contrast value, based on the point cloud Selecting a CT slice that best represents the characteristics of the archery among the plurality of CT slices constituting the 3D dental CT image as a reference CT slice, and placing a plurality of main control points on the arch form shown in the reference CT slice, Generating the curve along a major control point of.

The present invention has the effect of improving the efficiency of the overall dental diagnosis by automatically selecting the ideal slice for the curve to be generated from the three-dimensional dental CT image consisting of a plurality of CT slices.

The present invention has the effect of automatically generating curves from selected CT slices, thereby reducing the time and effort to obtain accurate results in using diagnostic software and devices.

1A is a diagram illustrating a curve input by a conventional user directly to a dental CT slice.
FIG. 1B is a view illustrating a longitudinal section of a tooth orthogonal to a curve directly input by a conventional user.
FIG. 1C is a view illustrating a panoramic image in which a vertical section is reconstructed along a curve directly input by a conventional user.
2 is a schematic flowchart of a method for automatically generating a curve corresponding to a arch form on an automatically selected tooth CT slice from a 3D CT image according to an embodiment of the present invention.
3 is a flowchart illustrating a point cloud generation step S100 according to an embodiment of the present invention.
4A is a diagram illustrating a histogram of a 3D dental CT image according to an embodiment of the present invention.
4B is a diagram illustrating a smoothed histogram curve according to an embodiment of the present invention.
FIG. 4C is a diagram illustrating a minimum threshold contrast value of a 3D dental CT image in a graph obtained by differentiating a histogram curve according to an embodiment of the present invention.
5 illustrates a point cloud generated based on a minimum threshold contrast value according to an embodiment of the present invention.
6 is a flowchart illustrating a step (step S200) of selecting a reference CT slice according to an embodiment of the present invention.
FIG. 7 is a diagram illustrating a triangle defined for triangle analysis in a method of automatically generating a curve corresponding to a arch form on an automatically selected tooth CT slice from a 3D CT image according to an embodiment of the present invention.
8 is a diagram illustrating a result of finding a CT slice having a minimum eigenvalue according to an embodiment of the present invention.
9 is a flowchart illustrating a step (step S300) of generating a curve from a selected reference CT slice according to an embodiment of the present invention.
10A through 10D are diagrams illustrating steps of generating major adjustment points on a selected reference CT slice according to an embodiment of the present invention.
FIG. 11 is a diagram for generating a curve by generating sub-adjustment points between major adjustment points generated on a selected reference CT slice according to an embodiment of the present invention.

The foregoing objects, features, and advantages will become more apparent from the following examples taken in conjunction with the accompanying drawings.

It is to be understood that the specific structure or functional description is illustrative only for the purpose of describing an embodiment according to the concept of the present invention and that the embodiments according to the concept of the present invention may be embodied in various forms, Should not be construed as limited to these.

The embodiments according to the concept of the present invention can make various changes and have various forms, so that specific embodiments are illustrated in the drawings and described in detail in the specification of the present application. However, it should be understood that the embodiments according to the concept of the present invention are not intended to limit the present invention to specific modes of operation, but include all changes, equivalents and alternatives included in the spirit and scope of the present invention.

Whenever an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but there may be other elements in between It should be understood. On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between. Other expressions for describing the relationship between components, such as "between" and "immediately between" or "adjacent to" and "directly adjacent to", should be interpreted as well.

The terminology used in the specification of the present application is used only to describe a specific embodiment and is not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. It is to be understood that the terms such as " comprises "or" having "in this specification are intended to specify the presence of stated features, integers, But do not preclude the presence or addition of steps, operations, elements, parts, or combinations thereof.

Unless defined otherwise, all terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the meaning of the context in the relevant art and, unless explicitly defined herein, are to be interpreted as ideal or overly formal Do not.

Hereinafter, the present invention will be described in detail with reference to the preferred embodiments of the present invention with reference to the accompanying drawings. Like reference symbols in the drawings denote like elements.

2 is a schematic flowchart of a method for automatically generating a curve corresponding to a arch form on an automatically selected tooth CT slice from a 3D CT image according to an embodiment of the present invention. As shown in FIG. 2, first, in step S100, a threshold contrast value for a voxel representing a tooth is set based on a 3D dental CT image input from the outside, and a point cloud is generated. do. Here, the threshold contrast value refers to a threshold value for the contrast value of the voxels constituting a specific portion of the 2D or 3D image. A point cloud refers to a list storing coordinates of voxels only for voxels having a contrast value greater than the threshold contrast value when a 2D or 3D image and a threshold contrast value are given.

In step S200, based on the point cloud, the CT slice that best represents the characteristics of the arch of the plurality of CT slices constituting the input 3D dental CT image is selected as the reference CT slice. Here, the CT slice that best represents the characteristics of the archery refers to a slice having the most correct tooth shape ('∩' shape) among CT slices containing teeth.

A schematic method of automatically selecting a reference CT slice is to generate an eigenvalue for each CT slice and to select a CT slice having the minimum of the eigenvalues for each CT slice. At this time, the eigenvalue is generated by extracting specific values from the image and combining them. Here, it is preferable to use a triangle analysis method to extract specific values from the image. Trigonometric analysis is based on the premise that the tooth's morphological structure is '∩', so that the highest point, the leftmost point, and the rightmost point are always present. It is a method of extracting specific values from an image by using properties of a reference triangle (for example, center and radius of circumscribed circle and inscribed circle, three cabinets, and three sides length).

In addition, a function that arithmetically combines features extracted from a CT slice into one unique information (unique value) is called an objective function. The objective function outputs the eigenvalues distinguished from other CT slices for the CT slice that the user wants to select. In this case, a minimization method for designing the output value of the objective function to be the minimum value may be used. The minimization method is mainly used when feature values in the CT slice that should not be selected are generally greater than the specific values in the reference CT slice that should be selected.

In addition, the objective function obtains weights in the learning phase of the algorithm and then uses the fixed weights in other CT images. In other words, the software trains the weights of the objective functions obtained from several CT images, and the same results are obtained even if the specific values obtained from the CT images not used in the training are used as inputs to the objective functions.

That is, even when a 3D CT image of any patient is input, an eigenvalue is calculated through an objective function of combining several specific values from the CT slice, and the CT slice having the minimum value among the calculated CT slice eigenvalues is the reference CT. The slice will be selected.

In operation S300, a plurality of main control points are disposed on the arch of the reference CT slice to generate the curve along the plurality of main control points. In step S300, a reference triangle set for each CT slice is used as a start step in the process of finding a main control point of the curve in step S200. Before creating the curve, setting the reference triangle first is because it reduces the search area in setting the exact position of the primary and secondary control points. This reduces the amount of computation until setting the final position of the adjustment points, which in turn improves the performance of the overall algorithm.

That is, a triangle similar to the shape of the patient's arch that extends from the patient's dental arch on the selected CT slice to the left and right molars based on the front teeth is defined, and a main control point of a predetermined length is set from one side of the triangle. The partial search region may be set for each control point, and the position of the tooth may be searched by a method similar to the divide and conquer for each inside of the partial search region. When the position of the main control points is determined, a curve is generated by filling a certain interval generated between the main control points with the sub adjustment points. Here, the divisional conquest method refers to a method of solving the present problem by dividing the problem into several partial problems having the same characteristics in order to solve any present problem.

In step S400, the tooth cross section is calculated along the generated curve and the panorama image is reconstructed.

Hereinafter, with reference to Figures 3 to 11, the detailed operation of each step of the tooth region CT slice selection and curve generation method will be described.

First, the threshold value of the contrast values of the voxels constituting a specific part of the CT image may be usefully used in the process of automatically selecting a reference CT slice. In relation to the intensity of the voxel, the portion of the CT slice that is stepped on has a high intensity value of the voxel, and the portion of the dark portion has a low intensity of the voxel. In CT images of the patient's teeth, empty areas or liquid parts generally have low contrast values, and dense parts such as bones and metal implants generally have high contrast values, and soft tissues such as skin and muscles It has a medium contrast value.

Therefore, the tooth, bone, and soft tissue can be specified from the CT image based on the intensity value of the voxel. In this case, range values may be used or a minimum threshold may be used. The range value is used in a method of setting a minimum value and a maximum value when specifying a part desired by the user, and the minimum threshold value is used in a method of setting only a minimum value for specifying the part including other parts besides the part desired by the user.

In the present invention, there is a need for a minimum threshold that includes both teeth and bones and a minimum threshold that includes only teeth. Here, the minimum threshold including only teeth is used as a reference value when generating a point cloud, and the minimum threshold including both teeth and bones is used as a reference value when the main control point, which will be described later, is moved to a position where the tooth is likely to be located. It is used.

However, due to the radiation dose, sensor sensitivity, location (eg, temperature and humidity), and patient's health of the CT image, the threshold value may differ from image to image. Can be. In view of this difference in thresholds, a separate algorithm is needed to automatically set the minimum threshold.

3 is a flowchart illustrating a point cloud generation step S100 according to an embodiment of the present invention. In addition, Figure 4a is a view showing a histogram for a three-dimensional dental CT image according to an embodiment of the present invention, Figure 4b is a view showing a smoothed histogram curve according to an embodiment of the present invention, Figure 4c A graph showing a minimum threshold contrast value in a three-dimensional dental CT image in a graph obtained by differentiating a histogram curve according to an embodiment of the present invention. 5 illustrates point clouds generated based on a minimum threshold contrast value according to an embodiment of the present invention.

As shown in FIG. 3, in step S110, for each CT slice, normalization is performed such that contrast values for all voxels constituting the entire CT slice region are included in a predetermined range. As a process for convenience of internal operation, the difference between the intensity values of the voxels exists even after normalization. According to an embodiment of the present invention, in the 3D CT image, 0 is used as the contrast value of the lowest voxel, and 255 is used as the contrast value of the highest voxel. Real numbers).

In operation S120, a histogram is generated based on the normalized CT image. Here, referring to FIG. 3A, the histogram refers to a distribution of voxel contrast values for the entire area of the CT image, the X axis represents the contrast value of the voxel, and the Y axis represents the value of the corresponding voxel value in the entire image. The histogram of the dental CT image shows a statistically similar distribution, but the range of the Y-axis value may vary depending on the CT image.

In step S130, the histogram is analyzed to obtain a minimum threshold contrast value for teeth and a minimum threshold contrast value including teeth and bones. In this case, referring to FIG. 3B, before the histogram is analyzed, a signal is converted into a signal, and then a smoothing process is performed to remove noise of the signal and leave only a large feature value.

Next, among the vertices of the curve obtained by differentiating the smoothed histogram curve, a point corresponding to a condition set by the user is automatically selected as a threshold. That is, the threshold corresponding to the tooth and the threshold including both the bone and the tooth may be specified in this differential process. In the graph of FIG. 3C, the X axis represents the pixel intensity of the voxel, and the Y axis represents the slope. In this case, the valley value corresponding to the highest contrast value among the plurality of valley values shown in the differential curve is the minimum threshold contrast value of the voxel corresponding to the tooth, and among the multiple apex values shown in the differential curve. The apex value corresponding to the highest contrast value corresponds to the minimum threshold contrast value of the voxel containing teeth and bones.

In step S140, the point cloud is generated based on the minimum threshold contrast value. Referring to FIG. 5, when a point is displayed at all coordinates in a list of voxels having a contrast value greater than the minimum threshold contrast value of a voxel corresponding to a tooth, a cloud-like form is referred to as a cloud. . In the present invention, the point cloud is set as a list of vector values having (x, y) in two dimensions and (x, y, z) in three dimensions. In obtaining a feature value and a unique value, which will be described below, the point cloud indexes the points to be searched to remove unnecessary parts of the search area to reduce the time required for the calculation. Can be.

6 is a flowchart illustrating a step of selecting a reference CT slice according to an embodiment of the present invention (step S200), and FIG. 7 illustrates an image of an automatically selected tooth CT slice from a 3D CT image according to an embodiment of the present invention. Figure 3 shows a triangle defined for triangle analysis in the method for automatically generating a curve corresponding to the arch form. FIG. 8 is a diagram illustrating a result of finding a CT slice having a minimum eigenvalue according to an embodiment of the present invention. The left side corresponds to a coronal projection image, and the right side corresponds to each CT slice. A graph showing eigenvalues.

In FIG. 6, under the assumption that the arch of the input CT image is in the form of [∩], various attributes of the triangle in the CT slice can be converted into specific values through a triangular analysis from one CT cloud pointed cloud. have.

To this end, first, in step S210, a reference triangle is set for each CT slice, and a specific value of each CT slice is calculated through the reference triangle. Referring to FIG. 7, the reference triangle may be set by acquiring the uppermost point T, the leftmost point L, and the rightmost point R from the point cloud generated in step S140. Three points obtained in two dimensions are defined as vectors having coordinates of (x, y), respectively.

In this way, a reference triangle can be set in each CT slice from the patient's arch to the skull, and the feature is defined using the set properties of the reference triangle. At this time, there may be a CT slice that can not set the reference triangle. For example, when the reference triangle is disturbed due to noise caused by a metal accessory of the patient, there may be a case where the reference triangle cannot be set due to insufficient number of points in the point cloud. These CT slices can be excluded from the specific extraction and eigenvalue generation below.

In the present invention, the feature extracted from the CT slice is, firstly, the distance between the inscribed circle and the circumscribed circle of the reference triangle, and secondly, the number of voxels of the entire image of the number of voxels exceeding the minimum threshold contrast value representing the tooth. To ratio. This specificity stems from the anatomical structure of the human tooth.

When the reference triangle was defined for all CT slices, the length between the circumscribed center and the inscribed center of the reference triangle was the shortest in the CT slice where the tooth is present, and the lower part (chin part) or the upper part based on the crown area of the tooth. CT slices in the nasal (nasal) region have a longer distance between the circumscribed circle and the inscribed center than the CT slice near the tooth.

However, even if the CT slice in which there is a low probability does not exist, another condition is required because the reference triangle can meet the above conditions (shortening of the circumscribed center and inscribed center of the reference triangle). Accordingly, in order to increase the probability of selecting a slice in which a tooth exists, a ratio of the number of voxels of the entire image of the number of voxels exceeding the minimum threshold contrast value representing the teeth may be added as a feature. Here, the number of voxels exceeding the minimum threshold contrast value representing the teeth may be the number of voxels corresponding to the arch.

To calculate the distance between the inscribed circle and the circumscribed circle of the reference triangle as a specific value, calculate the distance between point T and point L (S _R ), the distance between point L and point R (S _T ), and the distance between point R and point T By defining the length of the three sides of the reference triangle. The distance between two points uses [Equation 1] which is commonly used in the Cartesian coordinate system.

는 [수학식 2]를 사용하여 계산한다. [수학식 2]는 삼각형의 세 점의 좌표를 사용하여 외접원의 중심을 구하는 공식이다. By defining the coordinates of the three points of the reference triangle and the length of the three sides, the center coordinates of the circumscribed circle and the inscribed circle can be calculated. Center of circumscribed circle

Is calculated using [Equation 2]. Equation 2 is a formula for finding the center of the circumscribed circle using the coordinates of three points of a triangle.

은 [수학식 3]을 사용하여 계산한다. [수학식 3]은 삼각형의 세 변의 길이를 이용하여 내접원의 중심 좌표를 구하는 공식이다. Also, the center of inscribed circle

Is calculated using [Equation 3]. Equation 3 is a formula for calculating the center coordinates of an inscribed circle using the lengths of three sides of a triangle.

Next, in step S220, predetermined weights are given to specific values for each CT slice, respectively, and calculated as eigenvalues. Specifically, after extracting specific values for each CT slice (two-dimensional image), using the specific values to make one unique value is called an objective function. At this time, the function value of the objective function uses an energy minimization method such that the optimal value (that is, the eigen value in the reference CT slice) becomes the minimum value. The energy minimization method is a method in which a unique value calculated through an objective function has a small value in a CT slice desired by a user and, on the contrary, becomes larger as it moves away from a desired CT slice. In addition, an energy maximization method may be used for the objective function.

In the present invention, the objective function may consist of a polynomial. If a total of n features exist, the objective function is defined as shown in [Equation 4].

Where w ₁ ,... , w _n is each specific value x ₁ ,. , means a weight multiplied by x _n . The specific value may vary depending on the training data for which the training was performed, and there is no absolute value. b is a bias used to collectively change the value of the objective function. In the present invention, rather than evaluating an absolute value, a relative evaluation for comparing eigenvalues between a plurality of CT slices is performed, so that b can be set to zero.

In addition, the weight value is set so that the eigenvalue of the ideal CT slice is minimum, and if the number of features is not large, the weight can be set manually. The process of setting weights and evaluation can be automated by using a machine learning method, and when there are a large number of CT images required for a specific number and learning, a deep learning method can also be used. Not specifically specified in the specification of the.

In operation S230, the CT slice having the minimum eigenvalue is automatically selected. Referring to the graph shown in FIG. 8, the X axis represents an eigenvalue and the Y axis represents each CT slice number. On the graph, CT slices with an eigenvalue of zero are actually infinite infinity (∞) and are excluded from the ideal CT slice candidate. According to the energy minimization method, the CT slice with the lowest eigenvalue is selected. As shown in the coronal projection image of FIG. 8, it can be seen that the CT slice having the smallest eigenvalue is located near the tooth.

9 is a flowchart illustrating a step (step S300) of generating a curve from a selected reference CT slice according to an embodiment of the present invention, and FIGS. 10A to 10D illustrate a selected reference CT slice image according to an embodiment of the present invention. 11 is a diagram illustrating steps of generating major adjustment points in FIG. 11 is a diagram of generating a curve by generating sub-adjustment points between major adjustment points generated on a selected reference CT slice according to an embodiment of the present invention.

As shown in FIG. 9, after the CT slice is automatically selected, a curve corresponding to the arch form is generated in the selected CT slice. In this case, a reference triangle and a point cloud for the selected CT slice are used. Both of these can be used to narrow the search space to search for the arch in the selected CT slice. By narrowing the search area, it is possible to reduce the amount of unnecessary computation and improve the computation speed.

In addition, a main control point and a secondary control point are used in the process of generating a curve corresponding to the arch form. The main coordination point means directly involved in finding the archery, and the sub coordination point fills the gaps between the main coordination points at regular intervals. The reason for classifying the control points into two is to minimize the generation of curves that are not desired by the user when a part of the teeth is missing or broken in the CT slice generating the curve, or the noise of the image itself.

In step S310, a predetermined number of main control points are positioned on the sides between the leftmost point and the center point of the reference triangle and on the sides between the rightmost point and the center point, and are defined as starting points of the main adjustment points. 10A and 10B, the number of points located on one side of the reference triangle is preferably defined as an interval between 8 mm and 12 mm based on the measured distance of the CT slice (actual length of the subject shown in the image). . If the value is smaller than the reference value, the distance between the main control points is too far, and the error increases.On the contrary, if the value is larger than the reference value, the inflection point where the curve is bent becomes too large and the two-dimensional and three-dimensional panorama images are distorted. This can happen.

As shown in FIG. 10C, in step S320, a partial search area for each main control point is defined. The shape of the partial search area may define a rectangular area around the first created control point. Ideally, the rectangle should be between 8mm and 12mm as defined by the length of one side, but the aspect ratio of the rectangle can be adjusted according to the characteristics of the input CT image.

As shown in FIG. 10D, in operation S330, the main adjustment point is moved to the coordinates of the portion where the tooth is most likely to be positioned in the partial search region. Here, the coordinates of the portion where the tooth is most likely to be located mean a mid-positioned point (MPP) of the point cloud in the partial search region. The coordinate of this point calculates the arithmetic mean value of all the point (CLP: Cloud Points) coordinates defined in the point cloud in the partial search area in the same manner as in [Equation 5].

If the above process is applied to all defined main control points, the main control points are positioned at the main positions of the arch as shown in FIG. 10D.

Next, as shown in FIG. 11, in step S34O, a sub-adjustment point is set between adjacent main control points among the plurality of main control points, and a curve of the arch shape is generated along the main control points and the set sub-adjustment points. do. Since the generation of the sub-adjustment point uses an algorithm well known to those skilled in the art, a detailed description thereof will be omitted.

In addition, in the present invention, a splining algorithm generally known to those skilled in the art may be used. Splicing is a method of smoothly connecting three or more points together, and a detailed description thereof is omitted.

Although the invention has been described with reference to the embodiments shown in the drawings, this is merely exemplary, and those skilled in the art will appreciate that various modifications and equivalent other embodiments are possible therefrom. . Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

Claims (9)

A method for automatically generating a curve corresponding to a arch form on an automatically selected tooth CT slice from a three-dimensional CT image,
Selecting a reference CT slice representing the arch of the plurality of CT slices constituting an externally input 3D dental CT image; and
Arranging a plurality of main control points on the arch form indicated in the reference CT slice to generate the curve along the plurality of main control points;
Automatic generation of curves corresponding to archery shapes.
The method of claim 1,
Selecting the reference CT slice,
A threshold contrast value for a voxel representing a tooth is set based on the input 3D tooth CT image, and only a voxel having a contrast value greater than the threshold contrast value is generated, and a point cloud storing coordinates of the voxels is generated. step;
Based on the point cloud, selecting a CT slice that best represents a feature of the archery from the plurality of CT slices as the reference CT slice;
Automatic generation of curves corresponding to archery shapes.

3. The method of claim 2,
Generating the point cloud,
Normalizing contrast values of all voxels of the 3D dental CT image within a predetermined range;
Generating a histogram curve representing a distribution of voxel numbers according to contrast values for the normalized 3D dental CT image;
Calculating a minimum threshold contrast value for the voxel representing the tooth based on the histogram curve; and
Generating the point cloud with voxels having a contrast value greater than the minimum threshold contrast value;
Automatic generation of curves corresponding to archery shapes.
The method of claim 3, wherein
Computing the minimum threshold contrast value,
After differentiating the histogram curve, a valley value corresponding to the highest contrast value among the plurality of valley values shown in the derivative curve is set as the minimum threshold contrast value.
Automatic generation of curves corresponding to archery shapes.
3. The method of claim 2,
Selecting a CT slice that best represents the characteristics of the archery as a reference CT slice,
A reference triangle is set by connecting the center point of each archery and the leftmost point and the rightmost point shown in the plurality of CT slices based on the point cloud, and between the center of the inscribed circle of the reference triangle and the center of the circumscribed circle of the reference triangle. Calculating a distance and extracting the distance as a specific value;
Assigning each of the plurality of CT slices predetermined weights to each of the specific values and calculating them as eigenvalues; and
Selecting a CT slice having the minimum calculated eigenvalue among the plurality of CT slices as the reference CT slice;
Automatic generation of curves corresponding to archery shapes.
The method of claim 5, wherein
When the specific value is extracted, the ratio of the number of voxels representing the arches of the CT slice to the number of voxels of the entire area of the CT slice is further considered.
Automatic generation of curves corresponding to archery shapes.
The method of claim 5, wherein
Generating the curve,
Positioning the plurality of main control points on the side between the leftmost point and the center point of the reference triangle and on the side between the rightmost point and the center point;
Defining a partial search region for each of the plurality of main control points;
Calculating an arithmetic mean value of the coordinates of all voxels generating the point cloud in the partial search area, setting the arithmetic mean value as a moving coordinate, and moving a main control point in the partial search area to the moving coordinates; Steps and
Generating a curve along each major control point;
Automatic generation of curves corresponding to archery shapes.
The method of claim 7, wherein
Generating curves along each of the main control points
A sub-adjustment point is set between adjacent main control points among the plurality of main control points, and a curve of a arch form is generated along the main control points and the set sub-adjustment points.
Automatic generation of curves corresponding to archery shapes.
The method of claim 1,
And reconstructing the panoramic image of the archery along the generated curve.
Automatic generation of curves corresponding to archery shapes.

Images