KR20200129633A - A Method and Apparatus for Modeling Average Orbital Shape in Head and Neck CT image by Using Statistical Shape Model - Google Patents

Abstract

A method of modeling an average shape of an orbital bone in a head and neck computed tomography image includes the steps of: dividing an extracted orbital bone from an input image; defining an inner wall region from the divided extracted orbital bone to divide the inner wall region into a first patch region and a second patch region; searching for the same point in learning data for the divided first and second patch regions; and generating an average orbital bone model, which is an average shape model of the orbital bone, using the average position of the same point in the learning data and dispersion degree. According to the present invention, it is possible to increase the efficiency and shorten operation time of a surgery.

Description

A Method and Apparatus for Modeling Average Orbital Shape in Head and Neck CT image by Using Statistical Shape Model}

The present invention relates to a method and apparatus for modeling an orbital bone in 3D computed tomography (CT), and more particularly, to a statistical shape in a 3D computed tomography (CT) image. It relates to an average shape modeling method and apparatus using a model.

The average orbital shape modeling in the head and neck computed tomography image can be used as an important preprocessing task for designing a standard shape, identifying the shape of the orbital bone for reconstruction of the fracture site in the orbital bone fracture, and making an implant using 3D printing technology. have.

However, in the computed tomography of the head and neck, the shape of the orbital bone differs from person to person, and in particular, the orbital bone at the fracture site is very thin and has a brightness value similar to that of the surrounding soft tissue, so it was difficult to predict its shape.

Therefore, in the past, a flat plate was used for surgery, which was not considered clinical part.However, in order to make an accurate shape suitable for individual orbits using such a flat plate, it is very cumbersome for the operator and the time required for orbital bone reconstruction is increased. There was.

The present invention discloses a method for modeling an orbital bone using a statistical shape model in a three-dimensional omputed tomography (CT) image according to an embodiment.

In particular, in manufacturing a plate for restoring the orbital bone at the fracture site, whose shape differs from person to person, is very thin, and has a brightness value similar to that of the surrounding soft tissue, using a statistical shape model, angles according to gender, height, and weight A method of generating an average model of the orbital bone by scaling and analyzing the patient's orbital bone and CT data is disclosed.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems that are not mentioned will be clearly understood by those skilled in the art from the following description.

According to an embodiment, a method for modeling an average orbital bone shape in a computed tomography (CT) image of the head and neck includes the steps of: segmenting an extracted orbital bone from an input image; Designating an inner wall region from the divided extracted orbital bone and dividing the inner wall region into a first patch region and a second patch region; Searching for the same point in the training data for the divided first and second patch areas; And generating an average orbital bone model using the average position of the same point searched for in the training data and the degree of variance, wherein the average orbital bone model includes an average orbital bone model generation step. .

In addition, according to another embodiment, the step of searching for the same point is performed through similarity analysis through a Principal Component Analysis (PCA) method or a Singular value decomposition method. It is to search for the same point.

Further, in another embodiment, the step of dividing the extracted orbital bone may include extracting an orbital region from the input image; Dividing the nasal region in the orbital region using a three-dimensional region growth technique; Performing a two-dimensional orbital bone segmentation through a two-dimensional region growth technique and hole filling in the orbital region in which the nasal cavity is divided; And performing a three-dimensional shape generation and post-processing on the divided two-dimensional orbital bone to divide the extracted orbital bone.

In another embodiment, the step of dividing the patch area may further include designating a first patch area and a second patch area having a specific shape for an orbital area where loss occurs through boundary setting.

In addition, in another embodiment, the learning data is classified based on body data including at least one of sex and age.

In addition, an apparatus for modeling an orbital bone according to an embodiment includes: one or more processors; And at least one memory in which instructions for causing the at least one processor to perform an operation are stored when executed by the at least one processor, and the operation performed by the at least one processor divides the extracted orbital bone from the input image. An operation to do; An operation of designating an inner wall region from the divided extracted orbital bone and dividing the inner wall region into a first patch region and a second patch region; An operation of searching for the same point in the training data for the divided first and second patch areas; And an operation of generating an average orbital bone model, which is an average shape model of the orbital bone, using the average position of the same point in the training data and the degree of dispersion thereof.

In addition to this, another method for implementing the present invention, another system and a computer program for executing the method, and a computer-readable recording medium for recording the computer program may be further provided.

According to the present invention as described above, it has various effects as follows.

According to an embodiment, a method and apparatus for modeling an average orbital bone shape in a head and neck image, by improving the diversity of manufacturing orbital bone plates required for orbital bone reconstruction through improved average orbital bone modeling, You can shorten the time. Furthermore, in manufacturing a plate of a bone structure having various brightness values in various computed tomography images targeting the human body, it is possible to improve the degree of alignment by modeling the bone structure in a similar manner.

1 is a conceptual diagram illustrating a method for modeling an orbital bone in a head and neck computed tomography image according to an exemplary embodiment.
FIG. 2 is a diagram illustrating the structure of an orbital bone in a head and neck computed tomography image according to an exemplary embodiment.
3 is a flowchart illustrating a method of modeling an orbital bone according to an exemplary embodiment.
4 is a flowchart illustrating a method of segmenting an extracted orbital bone according to an exemplary embodiment.
5 is a diagram illustrating an example of dividing a patch area according to an exemplary embodiment.
6 is a diagram for explaining an example of generating an average shape model (ie, an average orbital bone model) in a divided patch area according to an exemplary embodiment.
7 is a diagram schematically illustrating an internal configuration of an orbital bone modeling apparatus according to an exemplary embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments to be posted below, but may be implemented in a variety of different forms, and only these embodiments make the posting of the present invention complete, and common knowledge in the technical field to which the present invention pertains. It is provided to completely inform the scope of the invention to those who have, and the invention is only defined by the scope of the claims. The same reference numerals refer to the same components throughout the specification.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used as meanings that can be commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not interpreted ideally or excessively unless explicitly defined specifically.

The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used in the specification, “comprises” and/or “comprising” do not exclude the presence or addition of one or more other elements other than the mentioned elements.

In the present specification,'medical image data' refers to an image acquired of a human body through a photographing device.

In the present specification, the "photographing device" refers to a device that photographs medical image data. The'imaging device' may include various medical imaging devices, such as a device that photographs the inside of a human body to diagnose a disease, such as an X-ray or a computed tomography (CT) device.

In the present specification, the'head and neck image' means medical image data for the head and neck regions.

In the present specification, “input image” refers to image data of the head and neck of a patient for which an average orbital bone model is to be generated.

In the present specification, the'orbital bone' refers to a bone in a recessed part of the facial hair bone that accommodates the eyeball, appendages of the lacrimal gland, and the like. The'orbital bone' is composed of multiple orbital walls.

In the present specification, the term'extracted orbital bone' means acquired as a region corresponding to the orbit in medical image data.

In the present specification, the "orbital region" refers to an area including the orbital bone in medical image data. That is, the'orbital region' means an area extracted from medical image data for obtaining the extracted orbital bone.

In the present specification, the "average orbital model" refers to an average model of the orbital bones generated based on the extracted orbital bones.

1 is a conceptual diagram illustrating a method for modeling an orbital bone in a head and neck computed tomography image according to an exemplary embodiment.

Referring to FIG. 1, an orbital bone modeling system on a head and neck computer according to an embodiment may include an orbital bone modeling apparatus 100 and a photographing apparatus 120. In addition, the orbital bone modeling apparatus 100 and the photographing apparatus 120 may be connected to the network 110.

First, the photographing apparatus 120 according to an exemplary embodiment may photograph a head and neck image. The photographing apparatus 120 may generate training image data for machine learning or training data for generating an average shape model of the orbital bone.

The orbital bone modeling apparatus 100 according to an exemplary embodiment discloses a method of modeling an average orbital bone model using a statistical shape model from a head and neck computed tomography image captured by the photographing apparatus 120. In particular, in manufacturing a plate for restoring the orbital bone at the fracture site, whose shape is different from person to person, is very thin, and has a brightness value similar to that of the surrounding soft tissue, the orbital of each patient considering body data using a statistical shape model By scaling and analyzing bone and CT data, an average model of the orbital bone can be created.

The network 110 according to an embodiment may include a wired network as well as a wireless network, and all various means for transmitting an image captured by the imaging device 120 to the orbital bone modeling device 100 are included. Can be understood.

2 is a diagram for explaining the structure of an orbital bone in a head and neck computed tomography image according to an exemplary embodiment.

As shown in Fig. 2, the orbital wall (hereinafter referred to as the orbital wall) is an orbital lateral wall (201), an orbital roof (202), an orbital medial wall, 203), and an orbital floor (204). It is composed of a pyramidal bone structure of the slope, and protects the eye and nerves adjacent to the eye. In addition, the orbital wall is composed of cortical bones with high brightness values and trabecular and thin bones with low brightness values. On the other hand, the orbital fracture to be confirmed through the segmentation of the orbital bone is one of the most common fractures, and especially occurs frequently in the thin orbital inner wall 203 and the orbital floor 204. In other words, in cranio-maxillofacial surgery for reconstruction of the orbital wall following orbital bone fracture, modeling of the orbital bone is essential to support the ocular position and restore the volume and shape of the orbit. However, in the computed tomography of the head and neck, the shape of the orbital bone differs from person to person, and in particular, the orbital bone at the fracture site is very thin and has a brightness value similar to that of the surrounding soft tissue, so it was difficult to predict its shape.

In particular, in the past, a flat plate that was not considered clinical part was used for surgery.However, in order to make an accurate shape suitable for individual orbits using such a flat plate, it is very cumbersome for the operator and the time required for orbital bone reconstruction may increase. there was.

Accordingly, the orbital bone modeling method according to an embodiment can reduce the operation efficiency and operation time by improving the diversity of fabrication of the orbital bone plate required for orbital bone reconstruction through average shape modeling using a statistical shape model.

Hereinafter, an operation of the orbital bone modeling method and the apparatus 100 will be described in detail with reference to FIGS. 3 to 6.

3 is a flowchart illustrating a method of modeling an orbital bone according to an exemplary embodiment.

In step S300, the orbital bone modeling method according to an embodiment may divide the extracted orbital bone from the input image. For example, it is possible to automatically segment the extracted orbital bone from a computed tomography image based on machine-learned data using a neural network. In addition, in order to segment the orbital region in a state in which the orbital bone is partially collapsed, the orbital bone may be restored through a two-dimensional growth technique and hole filling after dividing the nasal region modeled in 3D in a computed tomography image. The detailed extraction orbital bone segmentation method will be described later in detail with reference to FIG. 4.

In step S310, the orbital bone modeling method according to an embodiment designates an inner wall area from the divided orbit area, and divides the inner wall area into a first patch area and a second patch area.

Here, the first patch area and the second patch area may be set to have the same size. A method of setting the first patch area and the second patch area in the orbital inner wall area will be described later with reference to FIG. 5.

In step S320, the orbital bone modeling method according to an exemplary embodiment searches for the same point in the training data for the divided first and second patch regions.

In the step of searching for the same point, the same point may be determined in the learning data through similarity analysis through a Principal Component Analysis (PCA) method or a Singular value decomposition method. In this case, the learning data may be classified on the basis of body data including at least one of gender and age (eg, by age group such as gender, infant, child, adult, or age group). That is, based on the assumption that the shape of the orbital region of a person having similar body data will be similar, the average shape of the orbital region, which will be described later, can be modeled through similarity analysis.

In step S330, the orbital bone modeling method according to an embodiment may generate an average orbital bone model, which is an average shape model of the orbital bone, using the average position of the same point in the training data and the degree of variance thereof. In addition, by generating a three-dimensional average shape model, it is possible to freely measure the size and angle of the orbital inner wall.

4 is a flowchart illustrating a method of segmenting an orbital bone extracted from medical image data according to an exemplary embodiment.

In step S400, the orbital bone segmentation method according to an embodiment extracts an orbital region. The orbital area may be designated by manually setting the volume of intent (VOI) by the user. In this case, a predetermined number of slices may be selected based on the shape information in order to obtain the VOI of the 3D image.

In step S410, the orbital bone segmentation method according to an embodiment may segment the nasal cavity region using a three-dimensional region growth technique. In this case, the orbital bone and the nasal cavity may be divided through threshold setting. In particular, it is possible to divide the orbital bone after segmenting the nasal cavity through a three-dimensional growth technique that applies a threshold value to a plurality of slices.

For example, when a threshold value is applied only to the nasal cavity, a minimum of -1000HU to a maximum of -200HU may be selected as the threshold. In addition, when a threshold value for a bone region is set, it may be selected from a minimum of 300HU to a maximum brightness value.

In step S420, the orbital bone segmentation method according to an embodiment may perform extraction orbital bone segmentation through a two-dimensional region growth technique and hole filling for the orbital region in which the nasal region is divided. For example, in a two-dimensional slice image of a temporarily divided orbital bone, not only an empty space of a part separated by a bone, but also a hole inside the orbital bone may be filled.

In step S430, the orbital bone segmentation method according to an embodiment may segment the final extracted orbital bone by generating a 3D shape from the segmented two-dimensional extracted orbital bone and performing post-processing. Here, in the step of performing the post-processing, the resolution of the restored orbital region may be adjusted by adjusting the number and intensity of smoting. For example, it may be set so that the more the finishing strength is, the more the finishing is.

5 is a diagram illustrating an example of dividing a patch area according to an exemplary embodiment.

Referring to FIG. 5, a first patch region 510 may be first selected for a region in the extracted orbital bone that needs to be restored. Next, a second patch area 520 having the same size as the first patch area may be selected. On the other hand, when the first patch area and the second patch area overlap, it can be distinguished by an arbitrary patch boundary. Meanwhile, the order in which the patch regions are set should be the same for each training data.

In addition, in an embodiment, a first patch area and a second patch area having a specific shape are designated for the orbital area where loss occurs through boundary setting. The specific shape may be designated as a shape similar to a circle.

6 is a diagram for explaining an example of generating an average shape model (ie, an average orbital bone model) in a divided patch area according to an exemplary embodiment. 6 illustrates an example of searching for the same point in training data and generating an average shape model using a Principal Component Analysis (PCA) method according to an embodiment.

Specifically, the main components 610 and 620 are extracted from the divided first patch area 510 and the second patch area 520 through analysis of the main components, respectively. Here, the main components 610 and 620 correspond to feature values extracted according to the main component analysis method.

In addition, for each major component, the same point is searched on the learning data 630 and the major component of the same point is obtained. Finally, the average shape model 640 for each patch area is generated using the average position of the main component values obtained from the training data 630 and the degree of variance thereof. Here, the average shape model may include the step of normalizing each patch area using the average position and dispersion degree of each major component. In addition, the learning data 630 may mean a plurality of clinical data collected for machine learning to perform an average shape model.

On the other hand, in addition to the main component analysis method shown in FIG. 6, various methods capable of performing similarity analysis such as singular value decomposition can be used to generate the average shape model. They are easy to understand.

7 is a diagram schematically showing an internal configuration of an orbital bone modeling apparatus 100 according to an exemplary embodiment.

The orbital bone modeling apparatus 100 is a device capable of performing a function to be described later, and may be configured with, for example, a server computer or a personal computer. In one embodiment, the orbital bone modeling apparatus 100 may include one or more processors 130 and/or one or more memories 140.

According to an embodiment, an operation performed by the one or more processors 130 may include an operation of segmenting an extracted orbital bone from an input image (eg, a head and neck image); Designating an inner wall area from the divided orbital area, and dividing the inner wall area into a first patch area and a second patch area; An operation of searching for the same point in the training data for the divided first and second patch areas; And an operation of generating an average shape model of the orbital region using the average position of the same point in the training data and the degree of variance thereof. In addition, in addition to the above-described operation, a person skilled in the art can easily understand that the orbital bone modeling apparatus 100 according to an embodiment can perform the orbital bone modeling method described in FIGS. 1 to 6.

In an embodiment, at least one of these components of the orbital bone modeling apparatus 100 may be omitted, or another component may be added to the orbital bone modeling apparatus 100. In addition, additionally or alternatively, some components may be integrated and implemented, or may be implemented as a singular or plural entity. At least some of the internal and external components of the orbital bone modeling apparatus 100 are each other through a bus, general purpose input/output (GPIO), serial peripheral interface (SPI), or mobile industry processor interface (MIPI). It is connected and can send and receive data and/or signals.

One or more memories 140 may store various types of data. Data stored in the memory 140 is data acquired, processed, or used by at least one component of the orbital bone modeling apparatus 100, and may include software (eg, a program). The memory 140 may include volatile and/or nonvolatile memory. One or more memories 140 may store instructions for causing one or more processors 130 to perform an operation when executed by one or more processors 130. In one embodiment, the one or more memories 140 may store personalized information for one or more users and/or recommendation information for one or more products. In the present disclosure, programs or commands are software stored in the memory 140, and various functions are applied so that an operating system, an application, and/or an application for controlling the resources of the orbital bone modeling apparatus 100 can utilize the resources of the device. It may include middleware and the like provided to.

The one or more processors 130 may control at least one component of the orbital bone modeling apparatus 100 connected to the processor 130 by driving software (eg, a program or command). In addition, the processor 130 may perform various operations related to the present disclosure, processing, data generation, and processing. In addition, the processor 130 may load data, etc. from the memory 140 or may store the data in the memory 140.

In an embodiment, the orbital bone modeling apparatus 100 may further include a communication interface (not shown). The communication interface may perform wireless or wired communication between the orbital bone modeling apparatus 100 and another server or another external device (eg, the photographing apparatus 120 ). For example, the communication interface is eMBB (enhanced mobile broadband), URLLC (Ultra Reliable Low-Latency Communications), MMTC (Massive Machine Type Communications), LTE (long-term evolution), LTE-A (LTE Advance), UMTS ( Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (code division multiple access), WCDMA (wideband CDMA), WiBro (Wireless Broadband), WiFi (wireless fidelity), Bluetooth (Bluetooth), NFC (near field) communication), a global positioning system (GPS), or a global navigation satellite system (GNSS).

Various embodiments of the orbital bone modeling apparatus 100 according to the present disclosure may be combined with each other. Each of the embodiments may be combined according to the number of cases, and an embodiment of the orbital bone modeling apparatus 100 made by combining also falls within the scope of the present disclosure. In addition, internal/external components of the orbital bone modeling apparatus 100 according to the present disclosure described above may be added, changed, replaced, or deleted according to embodiments. In addition, the internal/external components of the above-described orbital bone modeling apparatus 100 may be implemented as hardware components.

Meanwhile, the method according to an exemplary embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of the program instructions may include not only machine language codes such as those produced by a compiler but also high-level language codes that can be executed by a computer using an interpreter or the like.

So far, the present invention has been looked at around its preferred embodiments. Those of ordinary skill in the art to which the present invention pertains will be able to understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered from an illustrative point of view rather than a limiting point of view. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

100: orbital bone modeling device
110: network
120: photographing device

Claims (7)

In the method of modeling the average shape of the orbital bone in the head and neck computed tomography (CT),
Segmenting the extracted orbital bone from the input image;
Designating an inner wall region from the divided extracted orbital bone and dividing the inner wall region into a first patch region and a second patch region;
Searching for the same point in the training data for the divided first and second patch areas; And
A method comprising the step of generating an average orbital bone model, wherein the average orbital bone model is an average shape model of the orbital bone by using the average position of the same point searched for in the training data and the degree of dispersion thereof. . The method of claim 1, wherein the searching for the same point comprises:
The method of searching for the same point through similarity analysis through a Principal Component Analysis (PCA) method or a Singular value decomposition method. The method of claim 1, wherein the step of dividing the extracted orbital bone,
Extracting an orbital region from the input image;
Dividing the nasal region in the orbital region using a three-dimensional region growth technique;
Performing a two-dimensional orbital bone segmentation through a two-dimensional region growth technique and hole filling in the orbital region in which the nasal region is divided; And
And performing a three-dimensional shape generation and post-processing on the divided two-dimensional orbital bone to divide the extracted orbital bone. The method of claim 1, wherein dividing the patch area comprises:
The method further comprising the step of designating a first patch region and a second patch region having a specific shape for the orbital region where loss occurs through boundary setting. The method of claim 1, wherein the training data,
The method of claim 1, wherein the method is distinguished based on body data including at least one of sex and age. One or more processors; And
When executed by the one or more processors, the one or more processors include one or more memories storing instructions for performing an operation,
The operation performed by the one or more processors,
An operation of segmenting the extracted orbital bone from the input image;
An operation of designating an inner wall region from the divided extracted orbital bone and dividing the inner wall region into a first patch region and a second patch region;
An operation of searching for the same point in the training data for the divided first and second patch areas; And
An orbital bone modeling apparatus comprising an operation for generating an average orbital bone model, which is an average shape model of the orbital bone, using the average position of the same point in the training data and the degree of variance thereof. An orbital bone modeling program in a computed tomography image, which is combined with a computer that is hardware and stored in a medium to execute the method of any one of claims 1 to 5.

Images