KR102267488B1 - Apparatus And Method For Dental Image Registration - Google Patents

Abstract

The present invention relates to a first image matching unit that generates a first matched image by registering a first tooth image and a second tooth image, and a plurality of vertices included in an nth order (where n is an integer greater than or equal to 1). An error calculating unit for calculating a matching error value for the n-th matched image, a virtual image generating unit generating a virtual image by applying a matching error value to the nth-order matched image, and (n+1)th order matching the nth-order matched image and the virtual image Provided is a dental image matching device including a second image matching unit that generates a registered image.

Description

Apparatus And Method For Dental Image Registration

The present invention relates to a dental image registration apparatus and method, and more particularly, to a dental image registration apparatus and method capable of registering a tooth image capable of improving registration accuracy.

In computer vision, when a scene or an object is photographed from a different time or point of view, images according to different coordinate systems are obtained. Image registration refers to processing for transforming such different images and displaying them in one coordinate system.

Through such image registration, it is possible to check the correspondence between images obtained through different measurement methods.

In a dental surgical guide software, an image registration process between computed tomography (CT) image data and oral scan image data is generally performed before entering a dental implant planning stage.

The image registered through this image registration process becomes the basis for implant planning work that determines the safe and optimal implant placement position by identifying the bone tissue and neural tube position, so the accuracy of image registration is very important in the subsequent procedure. have

In the image registration method provided by the conventional medical software, a user manually inputs a point that is a reference point for image registration, and image registration is performed based on this. According to such a conventional image registration method, since the user selects the reference point by roughly judging by the eye, the result is very inaccurate, so that the user's manual operation process is inevitably followed after image registration. That is, the user modifies the matching result by changing the position of the point or reselecting the point. As described above, according to the prior art, a user spends a lot of time in image registration due to the continuous iterative process of registration and correction, and there is a problem in that a satisfactory result cannot be obtained as much as the time spent.

As another conventional method, there may be a method of acquiring an image including a marker for use as an intraoral registration standard, and registering an image obtained from a heterogeneous imaging device based on the marker in the image, but this is an image acquisition Since the process of performing marking for registration in the patient's oral cavity must be premised, it is cumbersome and causes inconvenience to the patient.

Since these conventional methods match the images by comparing the distances between all vertices included in the image, there is a problem in that the image matching speed is reduced, and the system load for comparing the distances between the vertices is increased.

Accordingly, there is a need for a method capable of automatically performing image registration at a high speed and with high accuracy without the hassle of using a separate marker or manual manipulation.

The present invention provides a dental image registration apparatus and method that can improve user convenience by automatically performing image registration with high accuracy, thereby reducing the time required for implant planning and improving the accuracy of implant planning. aim to

The technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned above can be clearly understood by those of ordinary skill in the art to which the present invention belongs from the description below. There will be.

In order to solve the above problems, the present invention provides a first image matching unit for generating a first matched image by registering a first tooth image and a second tooth image, and n (here, n is an integer greater than or equal to 1) order matching An error calculator that calculates a matching error value for a plurality of vertices included in an image, a virtual image generator that generates a virtual image by applying a matching error value to an nth-order matched image, and an nth-order matched image and a virtual image Provided is a dental image matching device including a second image matching unit configured to generate an (n+1)th-order registered image by registration.

Here, the second image matching unit may generate the (n+1)th-order registered image by repeating the registration up to a reference number of times.

Also, the second image matching unit may generate the (n+1)th-order registered image by repeating registration until the sum of the matching error values is equal to or less than the reference value.

Also, the virtual image generator may generate a virtual image by moving positions of a plurality of vertices included in the nth-order matched image by a matching error value.

In addition, the dental image matching apparatus of the present invention extracts a plurality of first feature points from among a plurality of vertices included in a first tooth image, and corresponds to a plurality of first feature points among a plurality of vertices included in a second tooth image. It may further include a feature extraction unit for extracting a plurality of second feature points.

Here, the first image matching unit may compare the distances between the plurality of first feature points and the second feature points to match the first tooth image and the second tooth image.

In addition, the dental image registration apparatus of the present invention extracts a plurality of third feature points from among a plurality of vertices included in an nth-order matched image, and includes a plurality of third feature points corresponding to a plurality of third feature points among a plurality of vertices included in the virtual image. It may further include a feature extraction unit for extracting the fourth feature point.

Here, the second image matching unit may match the nth-order matched image and the virtual image by comparing distances between the plurality of third and fourth feature points.

In addition, the dental image matching apparatus of the present invention may further include a display unit for displaying the matching error value by color in the nth-order matched image.

Also, the first tooth image and the second tooth image are different types of images, and may be any one of a CT image, an oral scan image, and a magnetic resonance image, respectively.

In addition, the present invention provides a step of generating a first registered image by registering a first tooth image and a second tooth image, and matching of a plurality of vertices included in an nth order (where n is an integer greater than or equal to 1) a registered image The steps of calculating an error value, generating a virtual image by applying a matching error value to the nth-order matched image, and generating an (n+1)th-order registered image by registering the nth-order matched image and the virtual image. It provides a tooth image registration method comprising the.

According to the present invention, there is an effect of improving user convenience by automatically performing image registration with high accuracy, thereby reducing the time required for implant planning and improving the accuracy of implant planning.

The effects obtainable in the present invention are not limited to the above-mentioned effects, and other effects not mentioned may be clearly understood by those of ordinary skill in the art to which the present invention belongs from the following description. will be.

1 is a view showing the main configuration of a dental image registration apparatus according to an embodiment of the present invention.
2 is a detailed block diagram of a control unit according to an embodiment of the present invention.
3 is a view for explaining a process of image registration by the first image matching unit according to an embodiment of the present invention.
FIG. 4 is a diagram for explaining a method for the first image matching unit to register first and second tooth images using an Iterative Closest Point (ICP) algorithm according to an embodiment of the present invention.
5 is a view showing the matching error values calculated by the error calculator according to the color on the display unit according to an embodiment of the present invention.
6 is a diagram exemplarily illustrating a virtual image generated by a virtual image generator according to an embodiment of the present invention.
7 is a diagram illustrating a registered image generated by the second image matching unit 154 according to an embodiment of the present invention.
8 is a view for explaining a tooth image registration method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in more detail with reference to the accompanying drawings. In this case, it should be noted that the same components in the accompanying drawings are indicated by the same reference numerals as much as possible. In addition, detailed descriptions of well-known functions and configurations that may obscure the gist of the present invention will be omitted.

In an embodiment of the present invention, each component may be composed of one or more sub-components, and the electrical, electronic, and mechanical functions performed by each component may include electronic circuits, integrated circuits and ASICs (Application Specific Integrated Circuits). ) may be implemented with various known elements or mechanical elements, and may be implemented separately or two or more may be integrated into one.

1 is a view showing the main configuration of a dental image registration apparatus according to an embodiment of the present invention.

1 , the dental image matching apparatus 100 according to an embodiment of the present invention includes a communication unit 110 , an input unit 120 , a display unit 130 , a memory 140 , and a control unit 150 . can be configured.

The communication unit 110 may transmit or receive dental image data by performing communication with an external device such as an image acquisition device (not shown) and a server (not shown). To this end, the communication unit 110 5G (5 ^th generation communication), LTE-A (long term evolution-advanced), LTE (long term evolution), Bluetooth, BLE (bluetooth low energe), NFC (near field communication), etc. of wireless communication, and wired communication such as cable communication may be performed.

The input unit 120 generates input data in response to a user's input of the dental image matching apparatus 100 . The input unit 120 includes at least one input means. The input unit 120 includes a keyboard, a keypad, a dome switch, a touch panel, a touch key, a mouse, a menu button, etc. may include.

The display unit 130 displays dental image data according to the operation of the dental image matching apparatus 100 . The display unit 130 may include a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, and a micro electro mechanical systems (MEMS) display. and electronic paper displays. The display unit 130 may be combined with the input unit 120 to be implemented as a touch screen.

The memory 140 stores operation programs of the dental image matching apparatus 100 . Specifically, the memory 140 may store an image registration-related algorithm such as an Iterative Closest Point (ICP) algorithm and a marching cube algorithm for extracting an isosurface from 3D image data. Also, the memory 140 may store a plurality of dental image data received from an image acquisition device or the like.

The controller 150 performs a series of operation processes for matching the tooth images.

2 is a detailed block diagram of a control unit according to an embodiment of the present invention, and FIG. 3 is a diagram for explaining a process of image registration by the first image matching unit according to an embodiment of the present invention.

As shown in FIG. 2 , the controller 150 of the dental image matching apparatus 100 according to an embodiment of the present invention includes a first image matching unit 151 , an error calculating unit 152 , and a virtual image generating unit ( 153) and the second image matching unit 154 may be included.

Referring to FIG. 3 , the first image matching unit 151 generates a first matched image c by registering a first tooth image a and a second tooth image b.

Here, the first tooth image and the second tooth image are different types of three-dimensional images having different coordinate systems or resolutions due to reasons such as being acquired through different imaging devices or acquired at different viewpoints, respectively. It may be any one of a computed tomography (CT) image, an oral scan image, and a magnetic resonance image (MRI).

3 shows a first tooth image (a) as an oral scan image, and a second tooth image (b) as a CT image.

Here, the CT image (b) includes information on internal tissues such as the crown (the upper side of the tooth that appears outside the gum) among the periodontal tissues in the oral cavity, the root (the lower side of the tooth that is hidden inside the gum as a part coupled with the alveolar bone), and the alveolar bone. are doing That is, in the CT image (b), information on teeth and alveolar bone may be clearly displayed, but information on gums may not be accurately displayed.

On the other hand, the oral scan image (a) includes the shape of the crown among the periodontal tissues in the oral cavity and the shape of the gum around the root, which is not clearly shown in the CT image (b). That is, in the oral scan image (b), information on the crown and gum of the tooth may be clearly displayed, but information on the root and alveolar bone inside the gum may not be clearly displayed.

Because implant surgery must consider both internal and external conditions, such as the placement and combination of alveolar bone and fixtures, the space between each tooth, and the appearance and placement angle of the crown (artificial tooth) according to the bone mass, bone density and distribution of the alveolar bone. Rather than using either a CT image (b) or an oral scan image (a), a more accurate and complete procedure can be provided when these images are matched and used. That is, it is possible to provide comprehensive information for implant surgery by combining the shape of the gum around the root and the shape information on the root and alveolar bone therein.

Meanwhile, although not shown in the drawings, the dental image registration apparatus 100 according to an embodiment of the present invention may further include a direction alignment unit (not shown) and a preprocessor (not shown).

Here, the direction aligning unit (not shown) aligns the first tooth image and the second tooth image so that the first tooth image and the second tooth image face the same direction prior to image registration.

In addition, the preprocessor (not shown) makes the resolutions of the first tooth image and the second tooth image identical by configuring the unit distance representing the object in the volume space of the first tooth image and the second tooth image to be the same. Then, the voxel information of the first tooth image and the second tooth image is converted into vertex information by using the Marching Cube Algorithm.

Here, since the marching cube algorithm is an algorithm for extracting an isosurface from 3D image data and is widely used in the image technology field, a detailed description thereof will be omitted.

The first image matching unit 151 according to an embodiment of the present invention may register the first and second tooth images using the ICP algorithm stored in the memory 140 .

To this end, the image matching apparatus 100 according to an embodiment of the present invention extracts a plurality of first feature points from among a plurality of vertices included in the first tooth image (a), and includes them in the second tooth image (b). The method may further include a feature extraction unit (not shown) for extracting a plurality of second feature points corresponding to the plurality of first feature points from among the plurality of vertices. Here, the first and second feature points may be vertices constituting the tooth.

The first image matching unit 151 overlaps the first and second tooth image data and compares the distances between the plurality of first feature points and the second feature points extracted by the feature extraction unit to obtain a first The tooth image (a) and the second tooth image (b) may be registered.

Here, the first image matching unit 151 registers the first tooth image (a) and the second tooth image (b) until the sum of all distances between the plurality of first and second feature points is equal to or less than the reference value. Can be repeated.

Here, the reference value may be preset by a user and may vary according to a target image matching accuracy. That is, the higher the target image matching accuracy, the smaller the reference value.

4 is a view for explaining a method for the first image matching unit to register the first and second tooth images using an Iterative Closest Point (ICP) algorithm according to an embodiment of the present invention.

Referring to FIG. 4 , when the distance between the first feature points s1 , s2 , and s3 and the second feature points d1 , d2 , and d3 is sufficiently small by repeating the matching process, the second feature points d1, d2, The distance l1, l2, l3 of the extension line extending from the plane tangent to d3) to the first feature points s1, s2, s3, and the vertical vector of the extension line and the second feature points d1, d2, d3 The matching process may be repeated to decrease the distance.

As described above, since the dental image matching apparatus 100 according to an embodiment of the present invention matches the images by comparing only the distances between feature points of the first and second teeth images, the first and second teeth images are Compared to image matching by comparing the distances between all included vertices, the image registration speed can be improved, and the system load for comparing the distances between all vertices can be minimized.

The error calculator 152 may calculate a matching error value for a plurality of vertices included in an nth-order matched image (where n is an integer greater than or equal to 1), and store it in a difference map.

5 is a view showing the matching error values calculated by the error calculator according to the color on the display unit according to an embodiment of the present invention.

Referring to FIG. 5 , the display unit 130 may classify and display a matching error value for each color on an nth-order matched image using a difference map.

Here, as the absolute value of the matching error value is lower, it means that the vertices included in the first and second tooth images overlap each other well, and it can be expressed that the degree of matching is high.

In addition, as the absolute value of the matching error value is higher, it means that the vertices included in the first and second tooth images overlap each other misalignedly, and it can be expressed that the degree of overlap is low.

That is, when the matching error value is 0, the degree of overlap is the highest, and it can be expressed that the degree of overlap decreases in proportion to the absolute value of the overlap error value.

For example, the degree of overlap refers to a degree to which the surface of the other image protrudes and sinks from the surface of any one of these images when the first and second tooth images are overlapped.

Here, when the surface of the other image protrudes from the surface of the one image, the overlap error value may have a positive value, and when the surface of the other image is depressed rather than the surface of the one image, the overlap error value may have a negative value.

Here, the overlap error value is displayed in color so that the operator can easily recognize the image registration process, and can intuitively determine the accuracy of the image registration result.

For example, referring to the color table for each overlapping error value (the right table of FIG. 5), the overlapping error value is negative in the portion where the surface of the oral scan image (a) protrudes outside the surface of the CT image (b). It is displayed in red because the surface of the oral scan image (a) is recessed into the surface of the CT image (b), and the overlap error value has a positive value and is displayed in blue. An error value of 0 may be displayed in green.

The virtual image generator 153 generates a virtual image by applying a matching error value to the nth-order matched image.

6 is a diagram exemplarily illustrating a virtual image generated by a virtual image generator according to an embodiment of the present invention.

Specifically, the virtual image generator 153 may generate a virtual image by moving positions of a plurality of vertices included in the nth-order matched image by a matching error value.

For example, as shown in FIGS. 5 and 6 , when the matching error value for a vertex is 0.041, the position of this vertex can be moved by 0.041, and other vertices are similarly positioned by the matching error value. Each of them can be moved to create a virtual image.

The second image matching unit 154 generates an (n+1)th-order registered image by matching the nth-order matched image and the virtual image.

Here, the above-described feature extraction unit (not shown) extracts a plurality of third feature points from among a plurality of vertices included in the nth-order matched image, and corresponds to a plurality of third feature points among a plurality of vertices included in the virtual image. A plurality of fourth feature points may be extracted.

Then, the second image matching unit 154 may match the nth-order matched image and the virtual image by comparing the distances between the plurality of third and fourth feature points. That is, the second image matching unit 154 may register the nth-order registered image and the virtual image by using the ICP algorithm like the first image matching unit 151 .

The second image matching unit 154 may generate an (n+1)th-order registered image by repeating the registration up to a reference number of times. Here, the reference number may be preset by the user and may vary depending on the target image matching accuracy. That is, since the image registration accuracy improves as the number of times of image registration is repeated, the reference number increases as the target image registration accuracy increases. However, it should be noted that if the reference number is set too small, the number of image matching increases, which slows down the system processing speed.

Alternatively, the second image matching unit 154 may generate a (n+1)th-order registered image by repeating registration until the sum of the matching error values for each vertex is equal to or less than the reference value. Here, the reference value may be preset by a user and may vary according to a target image matching accuracy. That is, since the image registration accuracy is improved as the reference value is closer to 0, the reference value increases as the target image registration accuracy increases. However, it should be noted that if the reference value is set too small, the number of image matching increases and the system processing speed is slowed down.

7 is a diagram illustrating a registered image generated by the second image matching unit 154 according to an embodiment of the present invention.

Comparing FIGS. 5 and 7 , the registered image (see FIG. 7 ) generated by the second image matching unit 154 (see FIG. 7 ) is larger than the registered image (see FIG. 5 ) generated by the first image matching unit 151 . It can be seen that the accuracy is high. That is, it can be seen that a green part indicating a relatively high degree of registration appears more in the matched image of FIG. 7 than in the matched image of FIG. 5 .

As such, the dental image registration apparatus 100 according to an embodiment of the present invention automatically performs image registration with high accuracy to enhance user convenience, thereby reducing the time required for implant planning and reducing implant planning. accuracy can be improved.

8 is a view for explaining a tooth image registration method according to an embodiment of the present invention.

Hereinafter, a tooth image registration method according to an embodiment of the present invention will be described, but the content overlapping with the above will be omitted.

Referring to FIG. 8 , in the tooth image registration method according to an embodiment of the present invention, first, a first tooth image and a second tooth image are registered to generate a first registration image. Next, a matching error value for a plurality of vertices included in the nth order (where n is an integer greater than or equal to 1) matched image is calculated. Next, a virtual image is generated by applying a matching error value to the nth-order matched image. Next, an (n+1)-order registered image is generated by registering the nth-order registered image and the virtual image.

By repeating this registration process up to a reference number of times, an (n+1)-order matched image can be generated. On the other hand, by repeating the registration process until the sum of the matching error values for each vertex becomes less than or equal to the reference value ( An n+1) order matched image may be generated.

As such, the dental image registration method according to an embodiment of the present invention improves user convenience by automatically performing image registration with high accuracy, thereby reducing the time required for implant planning and improving the accuracy of implant planning. can do.

On the other hand, the embodiments of the present invention disclosed in the present specification and drawings are merely presented as specific examples to easily explain the technical contents of the present invention and help the understanding of the present invention, and are not intended to limit the scope of the present invention. That is, it will be apparent to those of ordinary skill in the art to which the present invention pertains that other modified examples can be implemented based on the technical spirit of the present invention.

100: tooth image matching device 151: first image matching unit
110: communication unit 152: error calculation unit
120: input unit 153: virtual image generation unit
130: display unit 154: second image matching unit
140: memory
150: control unit

Claims (11)

a first image matching unit configured to generate a first matched image by registering the first tooth image and the second tooth image;
an error calculator configured to calculate a matching error value for a plurality of vertices included in an nth-order matched image (where n is an integer greater than or equal to 1);
a virtual image generator for generating a virtual image by applying the matching error value to the nth-order matched image; and
A second image matching unit generating an (n+1)th-order registered image by registering the nth-order registered image and the virtual image
A dental image matching device comprising a.
The method of claim 1,
The second image matching unit
Repeating registration up to a reference number of times to generate the (n+1)th order registered image
Tooth image matching device.
The method of claim 1,
The second image matching unit
Repeating registration until the sum of the matching error values is less than or equal to a reference value to generate the (n+1)th order registered image
Tooth image matching device.
The method of claim 1,
The virtual image generator
generating the virtual image by moving positions of a plurality of vertices included in the nth-order matched image by the matching error value
Tooth image matching device.
The method of claim 1,
Extracting a plurality of first feature points from among a plurality of vertices included in the first tooth image, and extracting a plurality of second feature points corresponding to a plurality of first feature points from among a plurality of vertices included in the second tooth image extractor
A tooth image matching device further comprising a.
6. The method of claim 5,
The first image matching unit
matching the first tooth image and the second tooth image by comparing the distances between the plurality of first feature points and the second feature points
Tooth image matching device.
The method of claim 1,
Extracting a plurality of third feature points from among a plurality of vertices included in the nth-order matched image, and extracting a plurality of fourth feature points corresponding to the plurality of third feature points from among a plurality of vertices included in the virtual image part
A tooth image matching device further comprising a.
8. The method of claim 7,
The second image matching unit
and comparing the distances between the plurality of third and fourth feature points to match the nth-order matched image and the virtual image.
Tooth image matching device.
The method of claim 1,
The dental image matching apparatus further comprising a display unit for displaying the matching error value by color on the n-th matched image.
The method of claim 1,
The first tooth image and the second tooth image are
It is a different type of image, and each one of a CT image, an oral scan image, and a magnetic resonance image.
Tooth image matching device.
generating, by a first image matching unit, a first matched image by matching the first tooth image and the second tooth image;
calculating, by an error calculator, matching error values for a plurality of vertices included in an nth-order matched image (where n is an integer greater than or equal to 1);
generating a virtual image by applying, by a virtual image generator, the matching error value to the nth-order matched image; and
generating an (n+1)th-order registered image by a second image matching unit registering the nth-order registered image and the virtual image
A tooth image registration method comprising a.

Images