KR102235196B1 - Apparatus for medical image segmentation and method for modeling using the same - Google Patents

Abstract

It relates to a medical image segmentation apparatus and a modeling method using the same, wherein the medical image segmentation apparatus detects three or more arbitrary points within a specific organ area from at least one cross section included in the medical image. Closed curve generator that connects to create one closed curve, selects three or more arbitrary points as movement start points, and moves a plurality of selected movement start points in the outer direction of the closed curve, and expands the closed curve outward. When the closed curve corresponds to a specific organ area, a control unit that extracts the image of the closed curve area, and the images of the closed curve area extracted from a plurality of different sections included in the medical image are continuously collected to generate a three-dimensional shape of a specific organ. It includes a modeling unit that provides a three-dimensional shape of the generated specific organ.

Description

A medical image segmentation device and a modeling method using the same {APPARATUS FOR MEDICAL IMAGE SEGMENTATION AND METHOD FOR MODELING USING THE SAME}

A medical image segmentation apparatus and a modeling method using the same are provided.

During a surgical operation, in order to check the target area or organ, the corresponding area can be identified in advance through a CT image. However, since the patient's location and orientation may be different from those of the CT image, the CT image may not be of great help in direct identification of the organ. In addition, when a patient receives an explanation of the operation from a medical staff before the operation, if the explanation is received through a CT picture, the majority of patients who are not specialized in image reading may have poor understanding and satisfaction of the explanation. By analyzing the patient's medical image and creating a three-dimensional organ model through 3D printing, the medical staff can more accurately prepare for surgery, and the patient can also accurately recognize the affected area and condition through his organ model. have.

Through 3D printing, not only the patient's organs, but also the patient-specific orthodontic device can be manufactured using a variety of materials more easily than the conventional orthodontic device.

Korean Patent Publication No. 2014-0008031 Korean Patent Publication No. 2012-0111871 Korean Patent Publication No. 10-2011-0115762 Japanese Patent Laid-Open Publication No. Hei 09-006961 Japanese Unexamined Patent Application Publication No. 2000-107183 Japanese Unexamined Patent Application Publication No. 2000-339452

A problem to be solved by an embodiment of the present invention is to provide a medical image segmentation apparatus for deriving a boundary line of a specific organ from a cross section of a medical image.

A problem to be solved by an embodiment of the present invention is to provide a method for modeling by stacking segmentation regions of a medical image.

In addition to the above tasks, it can be used to achieve other tasks not specifically mentioned.

The medical image segmentation apparatus according to an embodiment of the present invention includes a detection unit that detects three or more arbitrary points within a specific organ area in at least one cross section included in the medical image, and connects three or more arbitrary points to one or more. A closed curve generator that creates a closed curve of, selects three or more arbitrary points as movement start points, and moves a plurality of selected movement start points in a direction outside the closed curve to expand the closed curve in an outward direction, and an extended closed curve is specified. In response to the boundary line of the organ region, a control unit that extracts the image of the closed curve region, and the images of the closed curve region extracted from a plurality of different sections included in the medical image are continuously collected to generate and generate a three-dimensional shape of a specific organ. It includes a modeling unit that provides a three-dimensional shape of a specific organ.

The extension unit compares the length of a line segment connecting a first arbitrary point located on the closed curve with a second arbitrary point adjacent to a preset distance, and if the length of the line segment is greater than a preset distance, the midpoint of the line segment is added to the moving start point. You can choose.

The extension unit includes a brightness derivation unit for deriving the brightness of an image at a position of a moving start point or a moving complete point moved by a predetermined distance, and an area where the moving complete point is located by comparing the brightness of the image with a preset threshold A position estimation unit that estimates the position of one of the inside of the region, the boundary line of a specific organ region, and the outside of a specific organ region, and selects movement start points excluding fixed points located at the boundary line of a specific organ region in response to the position of the movement completion point. And a moving start point controller that controls a moving direction and whether or not the moving start point is moved so that the moving complete point to which the selected moving select points have moved is located at a boundary line of a specific organ area.

When the brightness of the image is greater than a predetermined threshold, the position estimating unit estimates that the position of the moving end point is located inside a specific organ area. If the brightness of the image is equal to a predetermined threshold, the position of the moving end point is the boundary line of the specific organ area. If the brightness of the image is less than a predetermined threshold value, it may be estimated that the position of the moving end point is located outside a specific organ area.

When an arbitrary point or a movement completion point is selected as a movement starting point, the movement start point control unit moves a predetermined distance outside the closed curve based on a vertical bisector and a line segment between two neighboring movement starting points from the movement starting point, and the midpoint of the line segment When this movement starting point is selected, if the brightness of the image at which the movement starting point is located is greater than a predetermined threshold based on a line segment between two neighboring points and a vertical bisector from the selected movement starting point, it moves a predetermined distance outside the closed curve, If the brightness of the image in which the movement start point is located is less than a predetermined threshold value, the image may be moved in a direction inside the closed curve by a predetermined distance.

In the modeling method of a medical image segmentation apparatus according to an embodiment of the present invention, the step of detecting three or more arbitrary points within a specific organ area from at least one cross section included in the medical image, Connecting to create a closed curve, comparing the length of a line segment connecting two neighboring points among three or more random points with a preset distance, selecting three or more random points as the starting point for movement, and the length of the segment If is greater than a preset distance, further selecting the midpoint of the line segment as a movement starting point, moving a plurality of movement starting points in a direction outside the closed curve to expand the closed curve outward, and the expanded closed curve is a boundary line of a specific organ area Corresponding to, extracting and providing an image of the closed curve region, and generating a three-dimensional shape of a specific organ by continuously collecting images of the closed curve region extracted from a plurality of different sections included in the medical image, and generating a specific organ And providing a three-dimensional shape of.

In a computer-readable recording medium storing a program in which a method for extracting a specific organ region from a cross section within a medical image through a medical image segmentation device according to an embodiment of the present invention is stored, the program is included in the medical image. A function of detecting three or more arbitrary points within a specific organ area in at least one cross section, a function of connecting three or more arbitrary points to create one closed curve, selecting three or more arbitrary points as the starting point of movement, and , A function of moving a plurality of selected movement start points in a direction outside the closed curve to expand the closed curve in an outward direction, a function of extracting an image of the closed curve region when the expanded closed curve corresponds to a boundary line of a specific organ region, and a medical image It is intended to realize a function of generating a three-dimensional shape of the specific organ by continuously collecting images of the closed curve region extracted from a plurality of different cross-sections included therein, and providing a three-dimensional shape of the generated specific organ.

According to an embodiment of the present invention, by extending from the inside of a specific organ to a specific organ boundary line based on the characteristics of a medical image, the specific organ boundary line may be more accurately extracted to derive a cross section of the organ.

In addition, one embodiment of the present invention provides an image modeled by stacking cross sections of specific organs derived from medical images, so that medical staff or patients can recognize the exact state of treatment of specific organs, It can help in the manufacture of orthodontic devices.

1 is a block diagram showing a medical image segmentation apparatus according to an embodiment of the present invention.
2 is a block diagram showing a closed curve expansion unit according to an embodiment of the present invention.
3 is a flowchart illustrating a modeling method of a medical image segmentation apparatus according to an embodiment of the present invention.
4 is a diagram showing step S320 according to an embodiment of the present invention.
5 is a view showing step S345 according to an embodiment of the present invention.
6 is an exemplary view showing an embodiment of expanding a closed curve using an arbitrary point as a moving start point according to an embodiment of the present invention.
7 is an exemplary view showing an example in which a closed curve is expanded by using the midpoint of a line segment as a moving start point according to an embodiment of the present invention.
8 is an exemplary diagram for explaining a process in which an enlarged closed curve coincides with a boundary of a specific organ according to an embodiment of the present invention.
9 is an exemplary diagram for explaining a method of implementing a boundary using characteristics of an image according to an embodiment of the present invention.
10 is an exemplary view showing a 3D modeling screen generated by stacking segmentation regions of a medical image according to an embodiment of the present invention.

With reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those of ordinary skill in the art can easily implement the present invention. The present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and the same reference numerals are used for the same or similar components throughout the specification. In addition, in the case of well-known technologies, detailed descriptions thereof will be omitted.

Throughout the specification, when a part "includes" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary.

Hereinafter, "random point" refers to a point initially detected inside a specific organ area to be extracted from a cross section of a medical image, and "moving start point" refers to a point selected to move based on a certain condition. In detail, the "moving start point" represents an arbitrary point or a midpoint of a line segment connecting an arbitrary point to be located at the boundary of a specific organ area.

Hereinafter, the “movement end point” refers to a point at which the movement start point moves a predetermined distance and completes the movement. Among the movement end points, a point that is not located at the boundary of a specific organ area may be selected as a “movement start point” again.

Hereinafter, the "fixed point" refers to a point at which the boundary of a specific organ area is reached based on a certain condition among the moving end points, and when it is determined as a fixed point, the corresponding point is not selected as a moving start point.

Hereinafter, a medical image segmentation apparatus will be described in detail with reference to FIGS. 1 and 2.

1 is a block diagram showing a medical image segmentation apparatus according to an embodiment of the present invention, and FIG. 2 is a block diagram showing a closed curve extension unit according to an embodiment of the present invention.

As shown in FIG. 1, the medical image segmentation apparatus 100 includes a detection unit 110, a closed curve generation unit 120, a closed curve expansion unit 130, a control unit 140, and a modeling unit 150.

First, the detection unit 110 detects a plurality of arbitrary points within a specific organ area in at least one cross section included in the medical image.

In this case, the detection unit 110 detects three or more arbitrary points in which the characteristics of the image have a predetermined range of characteristic values at each location. For example, the detection unit 110 may detect arbitrary points having a value within the brightness range of the image in the medical image based on the brightness range of the image of the specific organ region to be extracted.

In addition, the detection unit 110 may detect four or more arbitrary points at a location in which the brightness of an image is greater than a predetermined threshold value corresponding to a boundary line of a specific organ region. Here, the threshold value may represent a range of characteristic values of a certain image.

The closed curve generator 120 generates one closed curve by connecting a plurality of arbitrary points.

In this case, when the closed curve generator 120 initially generates the closed curve, all of the connection points forming the closed curve may have a characteristic value of an image larger than a threshold value. Accordingly, the closed curve generator 120 allows the inside of the initial closed curve to be located inside a specific organ area.

In addition, when a point included in the closed curve moves, the closed curve generator 120 may update the closed curve including the moved position.

Next, the closed curve extension unit 130 selects one movement start point included in the closed curve, moves the selected movement start point in a direction outside the closed curve, and expands the closed curve in an outward direction.

The closed curve expansion unit 130 repeatedly expands the closed curve by moving start points toward the area outside the closed curve so that the closed curve includes the entire area of a specific organ.

In more detail, as shown in FIG. 2, the closed curve extension unit 130 may include a brightness derivation unit 131, a position estimation unit 132, a selection unit 133, and a movement start point control unit 134. .

The brightness derivation unit 131 derives the brightness of an image at each point where a plurality of movement start points or a plurality of movement completion points are located.

The position estimating unit 132 compares the brightness of the derived image with a preset threshold value, and the region where the movement start point or the movement completion point is located is one of the inside of the specific organ area, the boundary line of the specific organ area, and the outside of the specific organ area. It can be estimated as

For example, if the brightness of the image is greater than a predetermined threshold, the position estimating unit 132 estimates that the location of the moving complete point is located inside a specific organ area, and if the brightness of the image is equal to a predetermined threshold, the moving complete point It can be estimated that the location of is located at the boundary of a specific organ area. In addition, when the brightness of the image is less than a predetermined threshold value, the position estimating unit 132 may estimate that the position of the movement completion point is located outside a specific organ area.

Such a position estimation configuration is set based on a characteristic in which an organ (intestine, tooth, bone) region has a relatively higher brightness value than other regions in the CT image.

The selection unit 133 selects a plurality of movement start points from midpoints of a line segment connecting an arbitrary point and an arbitrary adjacent point. In this case, the selection unit 133 may repeatedly select the movement start points except for the fixed point located at the boundary line of the specific organ region, corresponding to the position of the movement completion point at the previous point in time.

The movement start point control unit 134 controls the movement direction and movement of the movement start point so that the movement start point is located at a boundary line of a specific organ region in correspondence with the estimated movement start point position.

The movement start point controller 134 may set an outer direction or an inner direction of the closed curve based on the position of the movement start point. In addition, when the movement end point at the previous time point is located at the boundary line of the specific organ region, the movement start point control unit 134 sets the corresponding movement end point as a fixed point, and the movement end point at the previous time point is not located at the boundary line of the specific organ region. If you can set the move repeatedly.

Next, when the expanded closed curve corresponds to the boundary line of a specific organ region, the controller 140 extracts and provides an image of the closed curve region. The controller 140 may store images of the closed curve region extracted from a plurality of different cross sections included in the medical image in a database (not shown) by describing a predetermined order.

The modeling unit 150 may continuously collect images of a closed curve region extracted from a plurality of different cross-sections included in the medical image to generate and provide a three-dimensional shape.

As such, the medical image segmentation apparatus 100 detects a specific organ region in a tomography medical image by using a characteristic value of an image such as brightness in the medical image or a threshold value of this characteristic, and is output as a two-dimensional planar image. Three-dimensional modeling of organs can be implemented by stacking regions.

Meanwhile, the medical image segmentation apparatus 100 may be a server, a terminal, or a combination thereof.

The terminal is collectively referred to as a device having computational processing capability by having a memory and a processor, respectively. For example, there are personal computers, handheld computers, personal digital assistants (PDAs), mobile phones, smart devices, and tablets.

The server is a memory in which a plurality of modules are stored, and a processor that is connected to the memory and reacts to the plurality of modules and processes service information provided to the terminal or action information that controls the service information, communication It may include means, and UI (user interface) display means.

Memory is a device that stores information, and is non-volatile such as high-speed random access memory, magnetic disk storage, flash memory device, and other non-volatile solid-state memory devices. Various types of memories such as volatile memory may be included.

The communication means transmits and receives service information or action information to and from the terminal in real time.

The UI display means outputs service information or action information of the device in real time. The UI display means may be an independent device that directly or indirectly outputs or displays the UI, or may be a part of the device.

Hereinafter, a method of deriving and modeling a specific organ region from a cross section of a medical image using a medical image segmentation apparatus will be described in detail with reference to FIGS. 3 to 10.

3 is a flowchart illustrating a modeling method of a medical image segmentation apparatus according to an embodiment of the present invention.

As shown in FIG. 3, the medical image segmentation apparatus 100 detects a plurality of arbitrary points within a specific organ area in a cross section of a medical image (S310).

The medical image segmentation apparatus 100 detects three or more arbitrary points having a characteristic value range of a predetermined characteristic value or a value greater than a threshold value at each location. Here, the threshold value indicates a range of characteristic values of a certain image and indicates a value corresponding to a boundary value of a specific organ region.

For example, when the difference in brightness of the image is large based on the boundary area according to the characteristics of the image of a specific organ area, the medical image segmentation apparatus 100 may You can detect the point. The criteria for detecting such an arbitrary point can be easily changed and designed based on the characteristics of the medical image in the future. In addition, when detecting one arbitrary point, the medical image segmentation apparatus 100 may detect arbitrary points located within a predetermined distance from the corresponding arbitrary point and satisfying the same condition.

Next, the medical image segmentation apparatus 100 generates a closed curve connecting a plurality of arbitrary points (S315).

When generating a closed curve, the medical image segmentation apparatus 100 may connect arbitrary points using a straight line or a constant curvature.

When initially generating the closed curve, the medical image segmentation apparatus 100 generates the closed curve so that the characteristics of the image have a range of a predetermined characteristic value or a value greater than a threshold value at all points forming the closed curve.

In other words, the medical image segmentation apparatus 100 generates a closed curve by generating a connection point or a connection line to satisfy the same condition as a plurality of arbitrary points, so that the closed curve is located inside a specific organ area.

Next, the medical image segmentation apparatus 100 selects a plurality of movement start points including a plurality of arbitrary points (S320). In addition, the medical image segmentation apparatus 100 selects a plurality of movement starting points and sets the value of i to 0 in association with each movement starting point.

4 is a diagram showing step S320 according to an embodiment of the present invention.

As shown in FIG. 4, the medical image segmentation apparatus 100 calculates a line segment length of each arbitrary point and a neighboring point in order to select a plurality of movement start points, and compares it with a predetermined distance (S321).

For example, the medical image segmentation apparatus 100 may compare a length of a line segment connecting a first arbitrary point and a second arbitrary point adjacent to a predetermined length of 4 pixels.

In addition, when the length of the line segment is equal to or smaller than the preset length, the medical image segmentation apparatus 100 selects a plurality of arbitrary points as a movement start point (S322).

On the other hand, if the length of the line segment is larger than the preset length, the medical image segmentation apparatus 100 selects a plurality of arbitrary points and midpoints of the line segments connecting the arbitrary points and neighboring points as the movement start point (S323).

For example, if the distance between an arbitrary point and a neighboring point is 6 pixels and the predetermined distance is 4 pixels, the midpoint of the line segment connecting the two points is moved because the distance between the two points is larger than the predetermined distance. You can choose it as a starting point.

As such, the medical image segmentation apparatus 100 may select arbitrary points and midpoints of a line segment connecting the arbitrary points as a movement start point.

For example, when four random points are set closed curves, and all line segments connecting adjacent random points have a value greater than a preset distance, the medical image segmentation apparatus 100 By selecting the midpoint of the line segment connecting the points, you can select a total of eight moving start points.

In this way, the medical image segmentation apparatus 100 selects a moving start point other than an arbitrary point derived in step S210, thereby increasing the number of movable points on the closed curve, thereby expanding the closed curve so that it coincides with the boundary of a specific organ area. can do.

Next, the medical image segmentation apparatus 100 moves each movement start point by a predetermined distance (S325).

The medical image segmentation apparatus 100 moves each movement starting point in the same direction as a corresponding vertical bisector based on a vertical bisector and a line segment connecting two points on a closed curve adjacent to the movement starting point, respectively. In addition, the medical image segmentation apparatus 100 may increase i by n indicating a predetermined distance set in advance.

Next, the medical image segmentation apparatus 100 derives the brightness from the position of the movement completion point where the movement start point has completed movement, and checks whether the brightness and the threshold value are matched (S330).

The medical image segmentation apparatus 100 derives the brightness from the position of the movement completion point, and if the brightness and the threshold value do not coincide, the previously set value of i is a 1/2 distance of a line segment that sets the movement direction of each movement start point. Check if it is a value (S335).

Here, when the moving start point is an arbitrary point, the line segment that sets the moving direction of the moving start point means a line segment between the arbitrary point and two neighboring points. In addition, when the movement start point is the midpoint of a line segment connecting an arbitrary point and an adjacent arbitrary point, the line segment that sets the movement direction of the movement starting point means a line segment connecting an arbitrary point and an adjacent arbitrary point.

If the previously set value of i is not the 1/2 distance value of the line segment, the medical image segmentation apparatus 100 returns to step S325, and the movement completion point is selected as the movement start point again from the position where the movement is completed, and corresponds to a predetermined distance. It can move, and can be increased by n, which represents a predetermined distance set in advance to i. Further, the process of comparing the brightness of the image at the location of the moving complete point and the threshold value may be repeated until the value of i becomes a 1/2 distance value of the line segment for setting the moving direction of the moving start point.

In addition, when the value of i associated with the repetitively moved movement completion point is a 1/2 distance value of the line segment, the medical image segmentation apparatus 100 may update a closed curve connecting the movement completion point (S340).

In addition, the medical image segmentation apparatus 100 selects a plurality of movement start points including a plurality of movement completion points excluding a fixed point (S345).

5 is a view showing step S345 according to an embodiment of the present invention.

As shown in FIG. 5, the medical image segmentation apparatus 100 calculates a length of a line segment of a movement completion point and a neighboring movement completion point, and compares it with a predetermined distance (S346).

Here, the predetermined distance may have the same distance value as the predetermined distance in step S321, and may be set to have a different value according to the situation.

The medical image segmentation apparatus 100 uses a length of a line segment between a movement completion point including a fixed point and a neighboring movement completion point, and when the length of the line segment is equal to or less than a preset length, the medical image segmentation apparatus 100 Selects a plurality of movement completion points as movement start points (S347).

In this case, the medical image segmentation apparatus 100 may select a plurality of movement completion points excluding a fixed point as a movement start point, and may set a value of i to 0.

On the other hand, if the length of the line segment is greater than the preset length, the medical image segmentation apparatus 100 selects a plurality of movement completion points excluding the fixed point and midpoints of the line segments connecting the movement completion points and the neighboring points as the movement start point (S348). ).

For example, when a fixed point is not set, the medical image segmentation apparatus 100 may select a total of 16 moving start points by selecting midpoints of a line segment between the eight moving complete points and the moving complete points.

In addition, when the medical image segmentation apparatus 100 selects a plurality of movement start points, it returns to step S325 again.

Meanwhile, when the brightness of the image matches the threshold value at the position of the movement completion point in step S330 described above, the medical image segmentation apparatus 100 sets the movement completion point as a fixed point, and the movement completion point including the fixed point It is possible to update the closed curve connecting the (S350).

As a result of comparing the brightness of the image with a predetermined threshold value, the medical image segmentation apparatus 100 determines that the corresponding movement start point has reached the boundary of a specific organ area when the brightness of the image matches the predetermined threshold value.

Accordingly, when the brightness of the image at the position of the moving start point matches a predetermined threshold, the medical image segmentation apparatus 100 is fixed to the corresponding point and is set as a fixed point so as not to be reselected as the moving start point.

Hereinafter, a process in which the medical image segmentation apparatus 100 moves a movement starting point will be described in detail with reference to FIGS. 6 and 7.

6 is an exemplary view showing an embodiment of expanding a closed curve using an arbitrary point as a moving start point according to an embodiment of the present invention, and FIG. 7 is a moving start point of a midpoint of a line segment according to an embodiment of the present invention. It is an exemplary view showing an embodiment of expanding the closed curve by using.

As shown in FIG. 6, in the medical image segmentation apparatus 100, a movement start point 601 selected from an arbitrary point moves by a predetermined distance.

At this time, the medical image segmentation apparatus 100 assumes a line segment between the selected movement start point 501 and two adjacent points 502 and 505, and moves in the same direction as the vertical bisector based on the vertical bisector for the assumed line segment. The starting point 601 is moved.

At this time, when the movement start point 601 moves by a predetermined distance in units of 0.1 mm, the medical image segmentation apparatus 100 derives the brightness characteristic of the image at the movement completion point and matches the brightness and a preset threshold. Make sure you do it. The preset threshold includes a brightness value corresponding to a boundary line of a specific organ region, and this may represent a range of a predetermined brightness value.

At this time, the medical image segmentation apparatus 100 may move a predetermined distance again and recheck the brightness value if the brightness of the image does not match a preset threshold at the location of the movement start point 601 that has moved a predetermined distance.

In addition, the medical image segmentation apparatus 100 may move the selected movement start point 501 between two neighboring points 502 and 505 by one half distance (a) of a line segment every 0.1 mm to check.

For example, if the distance between two points (502, 505) is 6 pixels, the moving distance must be within 3 pixels, which is half the distance between the two points, and the moving start point 501 is i. It moves three times while increasing the value, and extracts the feature value of the image and the brightness of the image at the moved position and checks the threshold value.

In addition, the medical image segmentation apparatus 100 may update the closed curve including the movement starting point 501 ′ when the movement starting point 501 is moved by a half distance (a) of the line segment.

Meanwhile, as shown in FIG. 7, it is determined that the length of the starting point 602 adjacent to the movement starting point 601 is longer than a preset length, and the movement starting point 610 is selected as the focus.

Further, the medical image segmentation apparatus 100 moves the movement start point 510 by a predetermined distance and derives the brightness of the image from the movement completion point 510'.

In addition, when the brightness of the image does not match a preset threshold value at the location of the movement completion point 510 ′ moved by a predetermined distance, the medical image segmentation apparatus 100 may move a predetermined distance again and recheck the brightness value.

At this time, the medical image segmentation apparatus 100 repeatedly moves until the movement start point 510 moves by a half distance (b) of the line segment connecting the movement start point 501 and the neighboring movement start point 502 and completes the movement. You can check the brightness of the image at the point.

In addition, the medical image segmentation apparatus 100 may update the closed curve including the movement completion point 510 ′ when the movement starting point 510 moves to the position of the movement completion point 510 ′ moved by a distance b.

On the other hand, the medical image segmentation apparatus 100, if the brightness of the image at the position of the moving complete point coincides with a threshold value in the process of moving the moving complete point by a 1/2 distance value of the line segment that sets the moving direction of the moving start point. You can stop the movement and set the movement completion point as a fixed point.

On the other hand, the medical image segmentation apparatus 100 checks whether the brightness of the image matches the threshold value at the selected movement starting point when a part of the closed curve includes outside a specific organ region according to the updated closed curve. You can further check whether it is a value or a small value.

For example, when the brightness of the image at the start point of movement is greater than the threshold value, the medical image segmentation apparatus 100 may estimate that the image is located inside a specific organ area. If the brightness of the image at the start point of movement is less than the threshold value, It can be assumed that it is located outside the organ area.

Therefore, when the brightness of the image at the moving start point is greater than the threshold value, the medical image segmentation apparatus 100 moves in the direction outside the closed curve described above, and if the brightness of the image at the moving start point is less than the threshold value, it moves toward the inside of the closed curve. You can set the direction.

In other words, the medical image segmentation apparatus 100 may set a movement direction such that a movement start point faces a boundary line of a specific organ area.

Next, the medical image segmentation apparatus 100 checks whether the brightness of the image matches the threshold value at the positions of all points on the closed curve including the set fixed point (S355).

When the medical image segmentation apparatus 100 determines that the brightness of the image on the closed curve does not match the threshold value, it returns to step S345.

Meanwhile, the medical image segmentation apparatus 100 may determine that the corresponding closed curve coincides with a boundary line of a specific organ region if the brightness of the image matches the threshold value at all points on the closed curve.

Therefore, the medical image segmentation apparatus 100 extracts the closed curve image and stacks the extracted closed curve images from a plurality of different cross sections to generate a specific organ in a three-dimensional shape, and provides the corresponding three-dimensional shape (S360).

At this time, the medical image segmentation apparatus 100 stores the closed curve image extracted from one cross section in an interlocking database, and sequentially extracts the closed curve image from the medical image or includes specific location information based on the order or specific location information. You can save it.

8 is an exemplary diagram for explaining a process in which an enlarged closed curve coincides with a boundary of a specific organ according to an embodiment of the present invention.

Figure 8 (A), (B), (C), (D), (E), (F), while the initial closed curve is generated in the cross section of the medical image, the closed curve expanded through the expansion method of the present invention is specified. It represents the process of matching the boundaries of the organ area.

In this way, in the medical image segmentation apparatus 100, the initial closed curve in the image is formed inside a specific organ region, and may be repeatedly expanded to extend to the boundary line of the specific organ region.

Meanwhile, the medical image segmentation apparatus 100 may continuously collect closed curve regions extracted from a plurality of different cross-sections included in the medical image to generate and provide a three-dimensional shape of a specific organ.

In this case, the medical image segmentation apparatus 100 may generate and provide a three-dimensional shape of a specific organ by using an image obtained by extracting a boundary of a specific organ using features of the image in addition to the cross-sectional image obtained by extracting the closed curve region described above.

9 is an exemplary diagram for explaining a method of implementing a boundary using an image characteristic according to an embodiment of the present invention.

9 is an exemplary diagram of a medical image generated by applying a color to a specific organ to the medical device itself, and (B) is a cross-sectional image by extracting all points having a specific brightness value when clearly distinguished from surrounding tissues such as bones or teeth. Is an example of creating

As shown in FIG. 9, a three-dimensional shape may be generated by collecting a cross-sectional image derived by dividing a boundary of a specific organ region using brightness, hue, and saturation of a medical image.

The medical image segmentation apparatus 100 may continuously collect closed curve regions from a plurality of different cross-sections included in the medical image, and may replace each point of the collected closed curve region with a rectangular parallelepiped.

In this case, the rectangular parallelepiped may be composed of horizontal and vertical intervals of points and intervals between cross-sections.

In addition, each cross section of the medical image segmentation apparatus 100 may be configured as a continuous rectangular parallelepiped, and has a disk shape.

Accordingly, the medical image segmentation apparatus 100 may generate a three-dimensional shape of a specific organ by continuously stacking the disk shapes. As such, the specific organ can be represented as a continuous disk or a set of rectangular parallelepipeds.

10 is an exemplary view showing a 3D modeling screen generated by stacking segmentation regions of a medical image according to an embodiment of the present invention.

10A and 10B are exemplary views showing a three-dimensional modeling that is rotated 180 degrees based on the reference point 801. As shown in FIG. 8, it is possible to provide a 3D modeling in which a color for each organ area is applied differently by stacking each segmentation area of a medical image.

Meanwhile, a method of implementing the modeling method through lamination is not limited to a specific method, and various methods suitable for a situation may be used.

For example, since the three-dimensional shape realized through lamination is expressed as a set of rectangular parallelepipeds, if the part in contact with the adjacent rectangular parallelepiped in each cube is deleted from the organ three-dimensional shape, only the surface of the organ consisting of a rectangle remains. At this time, the rectangle of the surface is divided diagonally to form a triangle, and one of the 3D printing files, a stereolithography (STL) file, may be generated by using the vertices of all the triangles.

In this case, the medical image segmentation apparatus 100 may further perform an operation of smoothly processing the surface of the specific organ.

In this way, the medical image segmentation apparatus 100 may perform modeling by continuously replacing and transforming a plurality of different cross sections included in the medical image.

Accordingly, a specific organ region in a tomography medical image may be segmented and stacked using the medical image segmentation apparatus 100, thereby implementing 3D modeling of a specific organ.

A program for executing a method according to an embodiment of the present invention may be recorded on a computer-readable recording medium.

The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The medium may be specially designed and configured, or may be known to and usable by a person skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magnetic-optical media such as floppy disks, and ROM, RAM, flash memory, and the like. Hardware devices specially configured to store and execute the same program instructions are included. Here, the medium may be a transmission medium such as an optical or metal wire, a waveguide, etc. including a carrier wave that transmits a signal specifying a program command, a data structure, and the like. Examples of program instructions 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.

Although one preferred embodiment of the present invention has been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art using the basic concept of the present invention defined in the following claims are also provided. It belongs to the scope of the present invention.

100: medical image segmentation device 110: detection unit
120: closed curve generation unit 130: closed curve expansion unit
131: brightness derivation unit 132: position estimation unit
133: selection unit 134: movement start point control unit
140: control unit 150: modeling unit

Claims (11)

A detection unit that detects three or more arbitrary points within a specific organ area in at least one cross section included in the medical image,
A closed curve generator for generating one closed curve by connecting the three or more arbitrary points,
An expansion unit configured to select the three or more arbitrary points as a movement start point, and move the selected plurality of movement start points in a direction outside the closed curve to expand the closed curve in an outward direction,
When the expanded closed curve corresponds to the boundary line of the specific organ region, a control unit for extracting an image of the closed curve region, and
A modeling unit that continuously collects images of the closed curve region extracted from a plurality of different cross-sections included in the medical image to generate a three-dimensional shape of the specific organ, and provides a three-dimensional shape of the generated specific organ,
The extension part,
The length of a line segment connecting a first arbitrary point located on the closed curve and a second arbitrary point adjacent to the closed curve is compared with a preset distance, and if the length of the line segment is greater than a preset distance, the midpoint of the line segment is further increased as the movement starting point. Medical imaging segmentation device of choice.
delete In claim 1,
The extension part,
A brightness derivation unit for deriving the brightness of an image at the location of the moving start point or the moving complete point moved by a predetermined distance,
A position estimating unit that compares the brightness of the image with a preset threshold value and estimates that the region where the movement completion point is located is one of the inside of the specific organ region, the boundary line of the specific organ region, and the outside of the specific organ region,
A selection unit for selecting movement start points excluding a fixed point located at a boundary line of the specific organ region corresponding to the position of the movement completion point, and
A medical image segmentation apparatus comprising a movement starting point controller configured to control a movement direction and a movement of the movement starting point so that the movement completion point to which the selected movement starting points moved is located at a boundary line of the specific organ area.
In paragraph 3,
The position estimation unit,
If the brightness of the image is greater than a predetermined threshold, it is assumed that the location of the moving complete point is located inside the specific organ area. If the brightness of the image is equal to a predetermined threshold, the location of the moving complete point is the specific A medical image segmentation apparatus that estimates to be located at a boundary line of an organ region, and estimates that the position of the movement completion point is located outside the specific organ region when the brightness of the image is less than a predetermined threshold value.
In paragraph 3,
The extension part,
When the arbitrary point or the movement completion point is selected as the movement starting point, the movement is moved by a predetermined distance outside the closed curve based on a vertical bisector and a line segment between two adjacent movement starting points from the movement starting point,
When the midpoint of the line segment is selected as the movement start point, if the brightness of the image at which the movement start point is located is greater than a predetermined threshold based on the line segment and the vertical bisector, it moves by a predetermined distance outside the closed curve, and the When the brightness of the image at which the movement start point is located is less than a predetermined threshold value, the medical image segmentation apparatus moves in a direction inside the closed curve by a predetermined distance.
Detecting three or more arbitrary points within a specific organ area from at least one cross section included in the medical image,
Connecting the three or more arbitrary points to create one closed curve,
Comparing a length of a line segment connecting two neighboring points among the three or more arbitrary points and a preset distance,
Selecting the three or more arbitrary points as a moving start point, and if the length of the line segment is greater than a preset distance, further selecting the midpoint of the line segment as a moving start point
Moving a plurality of movement start points in a direction outside the closed curve to expand the closed curve in an outward direction,
If the expanded closed curve corresponds to the boundary line of the specific organ region, extracting and providing an image of the closed curve region, and
A medical image comprising the step of continuously collecting images of the closed curve region extracted from a plurality of different cross-sections included in the medical image to generate a three-dimensional shape of the specific organ, and providing the three-dimensional shape of the generated specific organ. Segmentation device modeling method.
In paragraph 6,
The expanding step,
Deriving the brightness of the image at the location of the moving start point or the moving complete point moved by a predetermined distance,
Comparing the brightness of the image with a preset threshold and estimating a location in which the movement completion point is located within the specific organ region, a boundary line of the specific organ region, and an outside of the specific organ region,
Selecting movement start points excluding a fixed point located at the boundary line of the specific organ region corresponding to the position of the movement completion point, and
And controlling a movement direction and a movement of the movement starting point so that the movement completion point to which the selected movement starting points are moved is located at a boundary line of the specific organ region.
In clause 7,
Selecting the movement starting points,
When the arbitrary point or the movement completion point is selected as the movement starting point, and the distance at which the movement starting point is repeatedly moved by the predetermined distance becomes one half of a line segment between the movement starting point and two adjacent movement starting points, Update a closed curve including a movement complete point, and select a movement start point from the updated closed curve,
When the midpoint of the line segment is selected as the movement starting point, and the distance at which the movement starting point has been repeatedly moved by a preset predetermined distance becomes one half of the line segment, a closed curve including a movement completion point is updated, and the updated A method of modeling a medical image segmentation apparatus for selecting a movement start point from the closed curve.
In clause 7,
The step of estimating as the one location,
If the brightness of the image is greater than a predetermined threshold, it is assumed that the location of the moving complete point is located inside the specific organ area. If the brightness of the image is equal to a predetermined threshold, the location of the moving complete point is the specific A method of modeling a medical image segmentation apparatus for estimating to be located at a boundary line of an organ region, and estimating that the position of the moving complete point is located outside the specific organ region when the brightness of the image is less than a predetermined threshold value.
In clause 7,
The step of controlling the movement direction and whether the movement start point is moved,
When the arbitrary point or the movement completion point is selected as the movement starting point, the movement is moved by a predetermined distance outside the closed curve based on a vertical bisector and a line segment between two neighboring movement starting points from the movement starting point,
When the midpoint of the line segment is selected as the movement start point, if the brightness of the image at which the movement start point is located is greater than a predetermined threshold based on the line segment and the vertical bisector, it moves by a predetermined distance outside the closed curve, and the A method for modeling a medical image segmentation apparatus in which, when the brightness of the image at which the movement start point is located is less than a predetermined threshold value, a predetermined distance is moved in a direction inside the closed curve.
In a computer-readable recording medium storing a program in which a method for extracting a specific organ region from a cross section in a medical image through a medical image segmentation device is stored,
The above program,
A function of detecting three or more arbitrary points within a specific organ area from at least one cross section included in the medical image,
The function of creating one closed curve by connecting the three or more arbitrary points,
A function of selecting the three or more arbitrary points as a moving start point and further selecting the midpoint of the line segment as a moving start point when the length of a line segment connecting two adjacent points among the three or more arbitrary points is greater than a preset distance,
A function of expanding the closed curve outward by moving a plurality of selected movement start points in a direction outside the closed curve,
When the expanded closed curve corresponds to the boundary line of the specific organ region, a function of extracting an image of the closed curve region, and
A function of continuously collecting images of the closed curve region extracted from a plurality of different cross sections included in the medical image to generate a three-dimensional shape of the specific organ, and to provide a three-dimensional shape of the generated specific organ
To realize, a computer-readable recording medium.

Images