KR102709062B1 - Method and apparatus for registering a CT image with a surface scan model - Google Patents

Abstract

A device for registering a CT image and a surface scan model is provided. The device may include a storage unit for storing the CT image and the surface scan model, a display unit for displaying the CT image and the surface scan model, an input interface for selecting at least one target point from the CT image and at least one source point corresponding to the at least one target point from the surface scan model, and an image processing unit. The image processing unit may be configured to correct the at least one source point based on the at least one target point, and to register (global registration) the CT image and the surface scan model based on the at least one target point and the at least one corrected source point corresponding to the at least one target point.

Description

Method and apparatus for registering a CT image with a surface scan model

The present invention relates to medical image processing, and more specifically, to a technique for aligning a three-dimensional CT image and a surface scan model.

When implant placement and dental surgery are performed, it is very important for the safety of the patient to precisely identify the tissue structure and nerve canal location of the relevant organ at the pre-planning stage. For this reason, dental surgery practitioners want to examine the internal and external characteristics of the teeth more closely, but currently, CT images cannot provide a clear resolution that satisfies the user's needs due to the limitations of reconstruction technology. To overcome this shortcoming of CT images, a technology has been proposed that matches/registers surface information images modeled in advance with high resolution to existing CT images.

To support matching/registration of CT images and surface information images, most dental surgical guide software supports image registration functions. In the field of image signal processing, image registration refers to a technique that matches two images of the same object taken from different viewpoints or at different times by expressing them in a single coordinate space so that the morphologically identical parts of the object overlap and correspond.

The image registration method provided by conventional dental medical software is based on the user manually entering corresponding points that serve as registration criteria and calculating the geometric relationship between the input corresponding points in the three-dimensional coordinate space. The disadvantage of this method is that it is inconvenient to use because it requires excessive concentration from the user's perspective to click the exact corresponding points in the images without error. In addition, since the shapes between objects, i.e., characteristic similarity, cannot be compared during the calculation process, there is also the disadvantage that it is difficult to obtain visually sufficiently aligned results.

Another matching method is to automatically detect the most characteristic key points of an object by applying an image feature extraction algorithm to obtain more accurate correspondence points than the aforementioned method. This method can generally obtain accurate results more easily than the aforementioned method if the shapes and features of the two objects to be compared are the same. However, in the case of actual dental medical image data, there is a lot of noise due to light reflection from the patient's filled metal artifact and the environment, and some occlusions or non-identical surface shapes may be included. In such a situation, the 'feature-based automatic key point extraction method' is likely to cause unintended errors.

Another matching method is the ICP (Iterative Closest Point) method. The ICP method is a method that gradually minimizes the error by repeatedly moving in the direction where the distance error between the distributions of two objects decreases. This method can cope with the occlusion problem mentioned above to some extent because it finds a direction that satisfies the majority of areas even if some parts of the data have different shapes, and parts that are judged to have completely different shapes are processed as outliers. However, due to this characteristic, it only returns results that satisfy the minimum value in a local area, not the minimum error in the global data, and there is a problem that the matching result depends on the initial starting position of the comparison object. In addition, when a large amount of data is input into the system, the distance error must be calculated through comparison between all corresponding points, which can take too much processing time.

The object of the present invention is to provide a method and device for matching a CT image and a surface scan model, which improve the inaccuracy of the matching result due to incorrect manual input of corresponding points by a user, the difficulty of feature-based key point detection caused by shape obscuration or partial difference, and the problem of optimization to a local minimum due to an inappropriate initial starting position.

Another object of the present invention is to provide a method and device for matching a CT image and a surface scan model, which enables faster processing and more accurate results than before when applied to an object having the same shape in whole or in part.

Another object of the present invention is to provide a method and device for matching CT images and surface scan models, which allow a certain degree of positional error when inputting corresponding points from the user's perspective, thereby reducing the user's cognitive load and providing convenience of use.

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

In one aspect, a device for registering a CT image and a surface scan model is provided. The device includes a storage unit for storing the CT image and the surface scan model, a display unit for displaying the CT image and the surface scan model, an input interface for selecting at least one target point from the CT image and at least one source point corresponding to the at least one target point from the surface scan model, and an image processing unit, wherein the image processing unit may be configured to correct the at least one source point based on the at least one target point, and to register (global registration) the CT image and the surface scan model based on the at least one target point and the at least one corrected source point corresponding to the at least one target point.

In one embodiment, the image processing unit may be further configured to define a first block of a selected size centered on the at least one target point, define a second block of the selected size centered on the at least one source point, perform local registration of the first block and the second block, and correct the at least one source point according to a result of the local registration.

In one embodiment, the image processing unit may be further configured to perform a function of correcting the position of the at least one source point based on a rotational movement state of the second block at which the similarity between the first block and the second block is maximized while rotating the position of the second block.

In one embodiment, the image processing unit may be further configured to perform a function of locally aligning the first block and the second block by comparing features of the first block with features of the second block.

In one embodiment, the features of the first block may include at least one of a surface point cloud of the first block, a normal perpendicular to the surface, and a curvature of the surface, and the features of the second block may include at least one of a surface point cloud of the second block, a normal perpendicular to the surface, and a curvature of the surface.

In one embodiment, the image processing unit may be further configured to perform a function of locally aligning the first block and the second block using an ICP (Iterative Closest Point) algorithm with features of the first block and features of the second block.

In one aspect, a method for aligning a CT image and a surface scan model is provided. The method may include the steps of: (i) displaying the CT image and the surface scan model on a display, (ii) selecting at least one target point from the CT image, (iii) selecting at least one source point corresponding to the at least one selected target point from the surface scan model, (iv) correcting the at least one source point based on the at least one target point, and (v) correcting the CT image and the surface scan model based on the at least one target point and the at least one corrected source point corresponding to the at least one target point.

In one embodiment, the step (iv) may include: (iv-i) defining a first block of a selected size centered on the at least one target point, (iv-ii) defining a second block of the selected size centered on the at least one source point, (iv-iii) locally aligning the first block and the second block, and (iv-iv) correcting the at least one source point according to a result of the local alignment.

In one embodiment, the step (iv-iii) may include the step of (iv-iii-i) correcting the position of the at least one source point based on a rotational movement state of the second block in which the similarity between the first block and the second block is maximized while rotating the position of the second block.

In one embodiment, the step (iv-iii) may include the step (iv-iii-ii) of locally aligning the first block and the second block by comparing features of the first block with features of the second block.

In one embodiment, the features of the first block may include at least one of a point cloud of a surface in the first block, a normal perpendicular to the surface, and a curvature of the surface, and the features of the second block may include at least one of a point cloud of a surface in the second block, a normal perpendicular to the surface, and a curvature of the surface.

In one embodiment, the step (iv-iii-ii) may include a step of locally aligning the first block and the second block using an ICP algorithm with features of the first block and features of the second block.

According to the method and device for matching a CT image and a surface scan model according to embodiments of the present invention, when applied to an object having the same shape in whole or in part, there is a technical effect in that processing can be performed at a faster speed than in the past and accurate results can be obtained.

In addition, according to the method and device for matching CT images and surface scan models according to embodiments of the present invention, there is a technical effect in that a certain degree of positional error is allowed when inputting corresponding points from the user's perspective, thereby reducing the user's cognitive load and providing convenience of use.

In addition, according to the method and device for matching a CT image and a surface scan model according to embodiments of the present invention, there is a technical effect in that accurate matching can be achieved through preprocessing by local matching even if some of the shapes of the objects are different.

FIG. 1 is a drawing illustrating one embodiment of the configuration of a device for aligning a CT image and a surface scan model according to the present invention.
FIG. 2 is a flowchart illustrating one embodiment of a method for aligning a CT image and a surface scan model according to the present invention.
FIG. 3 is a drawing showing a CT image and a surface scan model according to the present invention side by side.
FIG. 4 is a drawing illustrating a result of aligning a CT image and a surface scan model according to one embodiment of the present invention.

The advantages and features of the present invention and the method for achieving them will become clear with reference to the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and the present embodiments are provided only to make the disclosure of the present invention complete and to fully inform a person having ordinary skill in the art to which the present invention belongs of the scope of the invention, and the present invention is defined only by the scope of the claims.

The terms used in this specification are used only to describe specific embodiments and are not intended to limit the present invention. For example, a component expressed in the singular should be understood to include a concept of plural components unless the context clearly indicates only the singular. In addition, in the specification of the present invention, terms such as "comprise" or "have" are intended to specify that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, and the use of such terms does not exclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof. In addition, in the embodiments described in this specification, a "module" or a "part" may mean a functional part that performs at least one function or operation.

In addition, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms defined in commonly used dictionaries, such as those defined in common dictionaries, should be interpreted as having a meaning consistent with the meaning they have in the context of the relevant art, and shall not be interpreted in an idealized or overly formal sense unless explicitly defined in the specification of the present invention.

Hereinafter, embodiments of the present invention will be described in more detail with reference to the attached drawings. However, in the following description, specific descriptions of widely known functions or configurations will be omitted if there is a risk of unnecessarily obscuring the gist of the present invention.

FIG. 1 is a drawing illustrating one embodiment of the configuration of a device for aligning a CT image and a surface scan model according to the present invention.

As illustrated in FIG. 1, the device (100) may include an input interface (110), an image processing unit (120), a storage unit (130), and a display unit (140). The input interface (110) may be configured with hardware and software modules for inputting user commands to perform image processing according to various embodiments of the present invention. The input interface (110) may be used to input various necessary commands to the image processing unit (120), or to input image data such as a CT image acquired by dental tomography, a surface scan model (image) that precisely scans and expresses tooth surface information, etc., to the storage unit (130). The input interface (110) allows the user to select a plurality of target points in the teeth portion of the displayed CT image, and also select a plurality of source points corresponding to the plurality of target points in the teeth portion of the surface scan model, thereby enabling various image processing to be performed accordingly. In the present invention, each of the plurality of target points and one of the plurality of source points corresponding thereto are defined as forming a point pair. In one embodiment, the input interface (110) may include a keyboard, a keypad, a touchpad, a mouse, etc. of a computer, but the type of the input interface is not limited thereto. For example, the input interface (110) may include a graphic user interface (Graphical User Interface) that can be controlled using the input devices described above. The display unit (140) is for displaying various images, such as a CT image and a surface scan model according to one embodiment of the present invention, and images formed according to various embodiments of the present invention, and may include various display devices, such as an LCD display, an LED display, an AMOLED display, and a CRT display.

The storage unit (130) may be used to store various images, such as CT images, surface scan models, and aligned images generated by matching a CT image and a surface scan model by the device (100) of the present invention. The storage unit (130) may be used to store image data of intermediate results obtained by performing image processing according to various embodiments of the present invention, result image data obtained by performing image processing according to various embodiments of the present invention, and variable values required to perform image processing according to various embodiments of the present invention. In various embodiments, the storage unit (130) may store the various images described above in a DICOM (Digital Imaging and Communications in Medicine) format or a general image file format (BMP, JPEG, TIFF, etc.). The storage unit (130) may further store software/firmware required to implement the image processing unit (120). The storage unit (130) may be implemented as any one of a storage medium among a flash memory type, a hard disk type, a MultiMedia Card (MMC), a card-type memory (for example, an SD (Secure Digital) card or an XD (eXtream Digital) card, etc.), a RAM (Random Access Memory), a SRAM (Static Random Access Memory), a ROM (Read-Only Memory), an EEPROM (Electrically Erasable Programmable Read-Only Memory), a PROM (Programmable Read-Only Memory), a magnetic memory, a magnetic disk, and an optical disk. However, those skilled in the art will recognize that the implementation form of the storage unit (130) is not limited thereto.

The image processing unit (120) may be configured to perform a function of defining a first block of a selected size centered on the corresponding target point for each of a plurality of point pairs, defining a second block of a selected size centered on the corresponding source point, performing local registration of the first block and the second block, and correcting the source point according to the local registration result. The image processing unit (120) may be further configured to determine a rotational movement state of the second block that maximizes the similarity between the first block and the second block while rotating the position of the second block when performing local registration of the first block and the second block, and to correct the position of the source point based on the rotational movement state of the second block that maximizes the similarity between the first block and the second block. The local registration may be performed by comparing the features of the first block and the features of the second block. In one embodiment, the features of the first block may include at least one of a surface point cloud of the first block, a normal perpendicular to the surface, and a curvature of the surface. In one embodiment, the feature of the second block may include at least one of a point cloud of a surface in the second block, a normal perpendicular to the surface, and a curvature of the surface. In one embodiment, the image processing unit (120) may be further configured to locally align the first block and the second block using an ICP (Iterative Closest Point) algorithm with the feature of the first block and the feature of the second block.

The image processing unit (120) may be further configured to globally register the CT image and the surface scan model based on a plurality of target points and a plurality of corrected source points each corresponding to the plurality of target points.

The image processing unit (120) may be implemented, in terms of hardware, using at least one of Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field-Programmable Gate Arrays (FPGAs), processors, controllers, micro-controllers, and microprocessors. The image processing unit (120) may also be implemented as a firmware/software module executable on the aforementioned hardware platform. In this case, the firmware/software module may be implemented by one or more software applications written in an appropriate program language.

FIG. 2 is a drawing illustrating an embodiment of a flow chart exemplifying a method for aligning a CT image and a surface scan model according to the present invention. FIG. 3 is a drawing illustrating a CT image and a surface scan model side by side according to the present invention.

The present method starts from a step (S210) of displaying a CT image (310) and a surface scan model (320) side by side on a display unit (140). The CT image (310) may be a three-dimensional image acquired by a CT imaging device. The surface scan model (320) may be an image acquired by precisely scanning tooth surface information. In the illustrated embodiment, the CT image (310) and the surface scan model (320) are displayed side by side left and right, but they may also be displayed side by side above and below, or they may be displayed separately. In step (S215), a user selects a target point (312) in a tooth portion of the CT image (310). In step (S220), the user selects a source point (322) corresponding to the target point (312) selected in the tooth portion of the surface scan model (320). In the embodiment of the present invention, a certain degree of positional error is allowed when selecting and inputting the target point (312) and the source point (322), which reduces the user's cognitive load and provides convenience to the user. The target point (312) and the source point (322) can be used as probabilistic clues for the corresponding positions of the local alignment to be described below. In step S225, a first block of a selected size is defined centered on the target point (312). In step S235, a second block of the selected size is defined centered on the source point (322). The sizes of the first block and the second block can be determined based on a reliability probability according to a Euclidean distance (actual size, mm) based on the target point (312) and the source point (322), respectively.

In step (S240), the first block and the second block are locally aligned. In one embodiment, the local alignment can be achieved by determining the rotational movement state of the second block at which the similarity between the first block and the second block is maximized while rotating the position of the second block. The similarity between the first block and the second block can be calculated by comparing the features of the first block and the features of the second block. Here, the features of the first block can include at least one of a point cloud of the surface in the first block, a normal perpendicular to the surface, and a curvature of the surface, and the features of the second block can also include at least one of a point cloud of the surface in the second block, a normal perpendicular to the surface, and a curvature of the surface. In one embodiment, the first block and the second block can be locally aligned using the ICP algorithm with the features of the first block and the features of the second block. In step (S245), the source point (322) is corrected according to the local alignment result. The position of the source point (322) can be corrected based on the rotational movement state of the second block where the similarity between the first block and the second block is maximized. In step (S250), steps (S240) and (S245) are repeated multiple times while changing the sizes of the first block and the second block. In one embodiment, steps (S240) and (S245) can be repeated multiple times while gradually increasing the sizes of the first block and the second block. An advantage can be obtained by performing hierarchical matching by starting from a narrow block area and defining a convex area that gradually includes a wider range and repeating steps (S240) and (S245). Matching in a narrow block area is efficient in noise removal, while matching in a wide range can relatively ensure overall matching stability, so that the overall matching accuracy can be improved by considering all of these matching results. Therefore, the matching accuracy can be further improved by correcting the source point (322) by weighting and averaging multiple local matching results.

In step (S255), steps (S215) to (S245) are performed repeatedly multiple times. That is, the user selects target points (314) and source points (324) for the tooth portions of the CT image (310) and the surface scan model (320), respectively, and performs local alignment to correct the position of the source points (324). Similarly, the user selects target points (316) and source points (326) for the tooth portions of the CT image (310) and the surface scan model (320), respectively, and performs local alignment to correct the position of the source points (326). In this way, when all of the source points (322, 324, 326) are corrected, a step of globally aligning the CT image (310) and the surface scan model (320) based on the plurality of target points (312, 314, 316) and the plurality of corrected source points (322, 324, 326) corresponding to the plurality of target points (312, 314, 316) is performed (see step (S255)).

In the above embodiment, it has been described that local alignment is performed after selecting one target point and one source point, but it is also possible to select multiple target points (312, 314, 316) and multiple source points (322, 324, 326) simultaneously and sequentially and perform local alignment for each point pair sequentially. In addition, in the above embodiment, it has been described that three target points and three source points are selected and local alignment is performed for each point pair, but the accuracy of alignment can be further improved by selecting any number of target points and source points and performing local alignment for each point pair.

FIG. 4 is a drawing illustrating a result of matching a CT image and a surface scan model according to one embodiment of the present invention. Referring to the drawing, it can be seen that the overall matching accuracy is greatly improved as a result of integrating the local matching results.

In the embodiments disclosed herein, the arrangement of the illustrated components may vary depending on the environment or requirements in which the invention is implemented. For example, some components may be omitted or some components may be integrated and implemented as one. In addition, the arrangement order and connection of some components may be changed.

Although various embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and the above-described embodiments can be variously modified and implemented by those skilled in the art without departing from the gist of the present invention as claimed in the appended claims, and such modified embodiments should not be understood separately from the technical idea or scope of the present invention. Therefore, the technical scope of the present invention should be defined only by the appended claims.

110: Input interface
120: Image processing unit
130: Storage
140: Display section
310: CT image
312, 314, 316: Target points
320: Surface Scan Model
322, 324, 326: Source points

Claims (12)

As a device for aligning CT images and surface scan models,
A storage unit for storing the CT image and the surface scan model,
A display section for displaying the CT image and the surface scan model;
An input interface that allows selecting at least one target point from the CT image and selecting at least one source point corresponding to the at least one target point from the surface scan model, and
Including an image processing unit,
The image processing unit is configured to define a first block of a selected size centered on the at least one target point, define a second block of the selected size centered on the at least one source point, perform local registration of the first block and the second block, correct the at least one source point based on the at least one target point according to the local registration result, and perform global registration of the CT image and the surface scan model based on the at least one target point and the at least one corrected source point respectively corresponding to the at least one target point.
The alignment device is further configured so as to perform a function of correcting the position of the at least one source point based on the rotational movement state of the second block at which the similarity between the first block and the second block is maximized while rotating the position of the second block. delete delete In the first paragraph,
The image processing unit is further configured to perform a function of locally aligning the first block and the second block by comparing the features of the first block and the features of the second block. In paragraph 4,
The feature of the first block includes at least one of a surface point cloud of the first block, a normal line perpendicular to the surface, and a curvature of the surface.
A feature of the second block is an alignment device, wherein the feature comprises at least one of a point cloud of a surface in the second block, a normal line perpendicular to the surface, and a curvature of the surface. In paragraph 4,
The image processing unit is further configured to perform a function of locally aligning the first block and the second block using an ICP (Iterative Closest Point) algorithm with the features of the first block and the features of the second block. A method for aligning CT images and surface scan models,
(i) a step of displaying the CT image and the surface scan model on a display;
(ii) a step of selecting at least one target point from the CT image;
(iii) a step of selecting at least one source point corresponding to at least one selected target point in the surface scan model;
(iv) a step of correcting at least one source point based on at least one target point, and
(v) a step of aligning the CT image and the surface scan model based on the at least one target point and the at least one corrected source point each corresponding to the at least one target point;
The above step (iv) is,
(iv-i) a step of defining a first block of a selected size centered on at least one target point;
(iv-ii) a step of defining a second block of the selected size centered on at least one of the source points;
(iv-iii) a step of locally aligning the first block and the second block, and
(iv-iv) comprising a step of correcting at least one source point according to the local alignment result;
The above step (iv-iii) is,
(iv-iii-i) A matching method, comprising a step of correcting the position of the at least one source point based on a rotational movement state of the second block at which the similarity between the first block and the second block is maximized while rotating the position of the second block. delete delete In Article 7,
The above step (iv-iii) is,
(iv-iii-ii) A matching method, comprising the step of comparing the features of the first block and the features of the second block to locally match the first block and the second block. In Article 10,
The feature of the first block includes at least one of a point cloud of a surface in the first block, a normal perpendicular to the surface, and a curvature of the surface.
A method for matching, wherein the feature of the second block comprises at least one of a point cloud of a surface in the second block, a normal line perpendicular to the surface, and a curvature of the surface. In Article 10,
The above step (iv-iii-ii) is a matching method including a step of locally matching the first block and the second block using an ICP algorithm with features of the first block and features of the second block.

Images