Method and system for automatically measuring anatomical landmark

Abstract

The present invention relates to a method and a system for automatically measuring anatomical feature points. According to the present invention, the method comprises the steps of: receiving a medical image by a medical image input unit; receiving, by a landmark input unit, a plurality of pieces of landmark information from a user with respect to the inputted medical image; and applying, by a landmark measurement unit, the inputted landmark to a pre-constructed algorithm to generate landmark line segment information and landmark angle information, and may further comprise a step of storing, by the landmark measurement unit, at least one of the plurality of generated landmark line segment information and landmark angle information in a database. According to the present invention, work efficiency can be improved by extracting three-dimensional anatomical landmarks for surgical surgery (for example, fracture, artificial joint insertion, spinal surgery, orthognathic surgery, etc.) and automatically performing analysis to assist in easily establishing a surgical plan.

Description

Method for automatically measuring anatomical landmarks and system thereof

[0001] The present invention relates to an automatic measurement method for anatomical feature points and a system thereof, and more specifically, to an automatic measurement method for anatomical feature points and a system thereof that automatically calculates distances and angles between anatomical feature points in an input medical image.

[0002] Cephalometry is an example of a method for extracting anatomical features from the human body and performing measurements. Cephalometry analyzes the relationship between the teeth and bones of the human skull, a technique used to establish treatment plans in orthodontics and oral and maxillofacial surgery.

[0003] Cephalometric analysis has traditionally been performed based on X-ray images, but this has the drawback of being repetitive and time-consuming in extracting anatomical landmarks. Furthermore, computed tomography (CT) images have recently been used to perform cephalometric analysis. However, this method requires a complex process, such as creating a two-dimensional image using a three-dimensional image to extract anatomical landmarks, and then extracting the anatomical landmarks from the two-dimensional image to perform measurements. This increases the complexity of this process, making it difficult to quickly apply it in clinical settings.

[0004] Thus, according to the present invention, an automatic measurement method and system for anatomical feature points that automatically calculate distances and angles between anatomical feature points in an input medical image are provided.

[0005] According to an embodiment of the present invention for achieving such a technical task, a method for automatically measuring anatomical landmarks includes: a step in which a medical image input unit receives a medical image; a step in which a landmark input unit receives a plurality of landmark information from a user for the input medical image; and a step in which a landmark measurement unit applies the input landmark information to a pre-established algorithm to generate landmark line information and landmark angle information, and may further include a step in which the landmark measurement unit stores at least one of the generated plurality of landmark line information and landmark angle information in a database.

[0006] The step of generating the feature point segment information and the feature point angle information may include the step of selecting a plurality of anatomical feature points from the plurality of input feature points; the step of generating a plurality of feature point segment information using the plurality of anatomical feature points; and the step of selecting two feature point segments from the plurality of feature point segments and generating feature point angle information based on whether the same anatomical feature point exists in the two selected feature point segments.

[0007] The step of inputting the above multiple feature point information may receive the coordinates of the multiple feature points and the identifier of the feature point corresponding to each coordinate from the user.

[0008] The step of selecting the plurality of anatomical feature points may select different plurality of anatomical feature points using a repeating statement or a conditional statement in the coordinates of the plurality of feature points.

[0009] The step of generating the plurality of feature point segment information may include the step of matching and storing each coordinate and identifier for the anatomical feature points; the step of selecting the two anatomical feature points and calculating the length of the feature point segment connecting the two selected anatomical feature points; and the step of generating the feature point segment information by matching the length of the feature point segment and the identifier of the feature point segment.

[0010] The step of generating the feature point angle information may include the steps of: storing the coordinates and identifiers of two feature points corresponding to each feature point line segment; selecting the two feature point line segments and determining whether the same anatomical feature point exists in the two selected feature point line segments; generating vectors of the two feature point line segments respectively when the same anatomical feature point exists in the two feature point line segments and calculating an angle between the two feature point line segments using an inner product to generate the feature point angle; and generating the feature point angle information by matching the calculated feature point angle with an identifier of the feature point angle.

[0011] The step of storing in the above database may store at least one of the feature point segment information and the feature point angle information in the database in the form of a class object.

[0012] In another embodiment of the present invention, an automatic anatomical feature point measurement system may include a medical image input unit for receiving a medical image; a feature point input unit for receiving a plurality of feature point information from a user for the input medical image; and a feature point measurement unit for applying the input feature points to a pre-established algorithm to generate feature point segment information and feature point angle information.

[0013] In this way, according to the present invention, work efficiency can be improved by assisting in easily establishing a surgical plan by extracting and automatically performing analysis of three-dimensional anatomical features for surgical operations (e.g., fracture, artificial joint insertion, spinal surgery, orthognathic surgery, etc.).

[0014] In addition, by creating feature point segment information and feature point angle information in the form of a class object, data post-processing such as data reconstruction and visualization can be facilitated.

[0015] Figure 1 is a configuration diagram of an automatic anatomical feature point measurement system according to one embodiment of the present invention.

[0016] Figure 2 is a flowchart of an automatic measurement method for anatomical landmarks according to another embodiment of the present invention.

[0017] Figure 3 is a flowchart of an algorithm according to another embodiment of the present invention.

[0018] Figures 4 to 6 are drawings showing examples of anatomical feature point measurement results.

[0019] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. In this process, the thickness of lines and the sizes of components depicted in the drawings may be exaggerated for clarity and convenience of explanation.

[0020] Throughout the specification, whenever a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise stated.

[0021] Furthermore, the terms described below are defined based on their functions within the present invention, and may vary depending on the intent or custom of the user or operator. Therefore, the definitions of these terms should be based on the overall content of this specification.

[0022] In the embodiment described below, the automatic measurement system for anatomical feature points (100) automatically calculates distances and angles between multiple feature points by applying a rule-based algorithm to multiple input feature points to automatically calculate distance and angle information between anatomical feature points, thereby automatically analyzing anatomical feature points, as described with a specific example.

[0023] Figure 1 is a configuration diagram of an automatic anatomical feature point measurement system according to one embodiment of the present invention.

[0024] As illustrated in (a) of FIG. 1, the anatomical feature point automatic measurement system (100) may include a medical image input unit (110), a feature point input unit (120), and a feature point measurement unit (130).

[0025] First, the medical image input unit (110) can receive medical images. Here, the medical images may be X-rays, ultrasound, computerized tomography (CT), magnetic resonance imaging (MRI), etc.

[0026] Specifically, the medical image input unit (110) can receive medical images from a picture archiving and communication system (PACS) or be directly connected to an image diagnosis device to receive medical images captured by the image diagnosis device.

[0027] Next, the feature point input unit (120) can receive multiple feature point information from the user for the input medical image.

[0028] Specifically, the feature point input unit (120) can receive information on multiple feature points of the input medical image through multiple feature point selection signals from the user for the input medical image. At this time, the multiple feature points include the three-dimensional coordinates and identifiers of each feature point.

[0029] Next, the feature point measurement unit (130) can apply the input feature points to a pre-built algorithm to generate feature point segment information and feature point angle information. Here, the algorithm can be built to select two anatomical feature points from among the plurality of input feature points to generate feature point segment information, and to select two feature point segments from among the plurality of generated feature point line segments to generate feature point angle information.

[0030] Additionally, the feature point measurement unit (130) can store at least one of the generated feature point segment information and feature point angle information in the database (200) in the form of a class object.

[0031] As shown in (b) of FIG. 1, the database (200) may be configured to include a feature point coordinate database (210), a feature point segment database (220), and a feature point angle database (230).

[0032] Specifically, the feature point coordinate database (210) can store coordinates for all input feature points.

[0033] Additionally, the feature point segment database (220) can store coordinates of anatomical feature points of feature point segments and their corresponding identifiers, and the length of feature point segments and their corresponding identifiers, in a matching manner.

[0034] Additionally, the feature point angle database (230) can store two feature point line segments, feature point angles generated using the same, and corresponding identifiers matched therewith.

[0035] The process of the feature point measurement unit (130) generating feature point segment information and feature point angle information and storing them in the database (200) will be described below with reference to FIGS. 2 to 6.

[0036] Hereinafter, with reference to FIGS. 2 to 6, a method for automatically measuring anatomical feature points according to an embodiment of the present invention will be described in more detail.

[0037] Figure 2 is a flowchart of an automatic measurement method for anatomical landmarks according to another embodiment of the present invention.

[0038] Referring to FIG. 2, the medical image input unit (110) can receive medical images (S210). Here, the medical images may be X-rays, ultrasound, computed tomography, magnetic resonance images, etc.

[0039] Specifically, the medical image input unit (110) can receive medical images from a medical image storage and transmission system or be directly connected to an image diagnosis device and receive medical images captured by the image diagnosis device.

[0040] Next, the feature point input unit (120) can receive multiple feature point information from the user for the input medical image (S220).

[0041] Specifically, the feature point input unit (120) can receive feature point information including a plurality of 3D feature point coordinates and an identifier of a feature point corresponding to each 3D feature point coordinate (e.g., first feature point, second feature point, …, nth feature point, etc.) from the user for the input medical image.

[0042] For example, the feature point input unit (120) can input coordinates (x1, y1, z1) and an identifier (e.g., a first feature point) from the user for the input medical image.

[0043] Next, the feature point measurement unit (130) can apply the input feature point to a pre-built algorithm to generate feature point segment information and feature point angle information (S230).

[0044] Figure 3 is a flowchart of an algorithm according to another embodiment of the present invention.

[0045] Referring to FIG. 3, the feature point measurement unit (130) can select multiple anatomical feature points from multiple input feature points (S310).

[0046] Specifically, the feature point measurement unit (130) can select a plurality of different anatomical feature points using a loop statement or a conditional statement. At this time, the loop statement may be, "Perform repeatedly until all feature point line segments connecting all feature points that do not overlap each other are generated" and "Perform repeatedly until the angles between all feature point line segments that do not overlap each other are generated.", and the conditional statement may be, "Select two feature points from the input feature points as anatomical feature points," "Select two feature point line segments that are in contact with each other in order to calculate an anatomical feature point angle," and "Select a feature point line segment that is in contact with each other based on whether or not the same anatomical feature point exists in the feature point line segments.".

[0047] For example, the feature point measurement unit (130) can select (x1, y1, z1), (x2, y1, z2), (x3,y3,z3), and (x4,y4,z4) from among a plurality of feature points as anatomical feature points.

[0048] Next, the feature point measurement unit (130) can generate feature point segment information using a plurality of anatomical feature points (S320).

[0049] Specifically, the feature point measurement unit (130) can match each coordinate of a plurality of anatomical feature points and the identifier of each anatomical feature point (e.g., first anatomical feature point, second anatomical feature point, feature point 1, feature point 2, etc.) and store them in the database (200) (S321).

[0050] For example, the feature point measurement unit (130) can match the coordinates and identifier of (x1, y1, z1) (e.g., a first anatomical feature point), the coordinates and identifier of (x2, y1, z2) (e.g., a second anatomical feature point), the coordinates and identifier of (x3, y3, z3) (e.g., a third anatomical feature point), and the coordinates and identifier of (x4, y4, z4) (e.g., a fourth anatomical feature point) and store them in the database (200).

[0051] In addition, the feature point measurement unit (130) can select two anatomical feature points to create a feature point line segment and calculate the length of the feature point line segment connecting the two selected anatomical feature points (S322).

[0052] For example, the feature point measurement unit (130) can select a first anatomical feature point and a second anatomical feature point and calculate the length of the feature point line segment connecting the first anatomical feature point and the second anatomical feature point.

[0053] In addition, the feature point measurement unit (130) can generate feature point segment information by matching the length of the feature point segment and the identifier of the feature point segment (e.g., first feature point segment, second feature point segment, feature point segment 1, feature point segment 2, etc.) for the generated feature point segment (S323).

[0054] For example, the feature point measurement unit (130) can generate a plurality of feature point segment information by matching the length of the feature point segment connecting the first anatomical feature point and the second anatomical feature point with an identifier (e.g., the first feature point segment).

[0055] Next, the feature point measurement unit (130) can select two feature point segments from among a plurality of feature point line segments and generate feature point angle information based on whether the same anatomical feature point exists in the two feature point line segments (S330).

[0056] Specifically, the feature point measurement unit (130) can match each coordinate and identifier for each anatomical feature point of a plurality of feature point line segments and store them in the database (200) (S331).

[0057] For example, the feature point measurement unit (130) can match the coordinates of the first feature point line segment and the first anatomical feature point and the coordinates of the second anatomical feature point and store them in the database (200).

[0058] In addition, the feature point measurement unit (130) can select two different feature point segments from among multiple feature point line segments using a repeat statement and a conditional statement, and determine whether the two selected feature point line segments have the same anatomical feature point (S332).

[0059] At this time, the feature point measurement unit (130) can select feature point segments using two nested loop statements and a conditional statement. Here, the first loop statement is set to repeat as many times as the total number of feature point segments, the second loop statement is set to repeat from the index of the first feature point segment to the last index, and all different feature point segments are selected through a conditional statement that selects feature point segments depending on whether the index of the first loop statement and the index of the second loop statement are different, and the feature point segments can be determined through a conditional statement that finally selects feature point segments for calculating feature point angles depending on whether two identical anatomical feature point coordinates exist among four anatomical feature point coordinates of two different feature point segments that have been selected.

[0060] In addition, if the same anatomical feature point exists in two feature point line segments, the feature point measurement unit (130) can generate a feature point angle by generating vectors of the two feature point line segments and calculating the angle between the two feature point line segments using the inner product (S333).

[0061] For example, the feature point measurement unit (130) selects a first feature point line segment and a second feature point line segment, and since the selected first feature point line segment and the second feature point line segment have the same anatomical feature point, vectors are generated for the first feature point line segment and the second feature point line segment, respectively, and an inner product is used to calculate the angle between the first feature point line segment and the second feature point line segment, thereby generating a feature point angle.

[0062] In addition, the feature point measurement unit (130) can generate feature point angle information by matching the calculated feature point angle with the identifier of the feature point angle (e.g., first feature point angle, second feature point angle, feature point angle 1, feature point angle 2, etc.) (S334).

[0063] Next, the feature point measurement unit (130) can store the generated multiple feature point segment information and feature point angle information in the database (200) (S240).

[0064] Specifically, the feature point measurement unit (130) can store at least one of the generated feature point segment information and feature point angle information in the database (200) in the form of a class object.

[0065] Figures 4 to 6 are drawings showing examples of anatomical feature point measurement results.

[0066] As illustrated in FIGS. 4 and 6, the anatomical feature point automatic measurement system (100) can extract 3D anatomical feature points for 3D medical images and analyze them by calculating the distance and angle between each anatomical feature point.

[0067] Furthermore, the anatomical landmark automatic measurement system (100) can combine analysis results with input medical images and provide them to the user.

[0068] In this way, according to an embodiment of the present invention, it is possible to improve work efficiency by assisting in easily establishing a surgical plan by extracting and automatically performing analysis of 3D anatomical features for surgical operations (e.g., fracture, artificial joint insertion, spinal surgery, orthognathic surgery, etc.).

[0069] In addition, by creating feature point segment information and feature point angle information in the form of a class object, data post-processing such as data reconstruction and visualization can be facilitated.

[0070] While the present invention has been described with reference to the embodiments illustrated in the drawings, these are merely exemplary, and those skilled in the art will understand that various modifications and equivalent embodiments are possible. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the following claims.

Claims

1. A step in which the medical image input unit receives medical images; A step in which a feature point input unit receives a plurality of feature point information from a user for the input medical image; and An automatic measurement method for anatomical feature points, comprising a step of a feature point measurement unit applying the input feature points to a pre-established algorithm to generate feature point segment information and feature point angle information.

2. In paragraph 1, An automatic measurement method for anatomical feature points, further comprising a step of storing information on a plurality of feature point segments and feature point angles generated by the feature point measurement unit in a database.

3. In paragraph 1, The step of generating the above feature point segment information and feature point angle information is: A step of selecting a plurality of anatomical feature points from the plurality of feature points input above; A step of generating a plurality of feature point segment information using the plurality of anatomical feature points; and An automatic measurement method for anatomical feature points, comprising the steps of selecting two feature point line segments from among the plurality of feature point line segments and generating feature point angle information based on whether the same anatomical feature point exists in the two selected feature point line segments.

4. In paragraph 1, The step of inputting the above multiple feature point information is: An automatic measurement method for anatomical feature points that receives coordinates of multiple feature points and feature point identifiers corresponding to each coordinate from a user.

5. In paragraph 3, The step of selecting the above multiple anatomical features is: An automatic measurement method for anatomical feature points that selects different anatomical feature points using a repeating statement or a conditional statement from the coordinates of the above plural feature points.

6. In paragraph 3, The step of generating the above multiple feature point segment information is: A step of matching and storing each coordinate and identifier for the above anatomical feature points; A step of selecting the two anatomical feature points and calculating the length of the feature point line connecting the two selected anatomical feature points; and An automatic measurement method for anatomical feature points, comprising a step of generating feature point segment information by matching the length of the feature point segment and the identifier of the feature point segment.

7. In paragraph 3, The step of generating the above feature point angle information is: A step of storing the coordinates and identifiers of two feature points corresponding to each feature point line segment; A step of selecting the two feature point line segments and determining whether the two selected feature point line segments have the same anatomical feature point; If the same anatomical feature point exists in the two feature point line segments, a step of generating vectors of the two feature point line segments and calculating the angle between the two feature point line segments using the inner product to generate a feature point angle; and An automatic measurement method for anatomical feature points, comprising a step of generating feature point angle information by matching the above-described feature point angle and the identifier of the feature point angle.

8. In paragraph 2, The step of saving to the above database is: An automatic measurement method for anatomical feature points, wherein at least one of the feature point segment information and feature point angle information is stored in a database in the form of a class object.

9. Medical image input unit for receiving medical images; A feature point input unit that receives multiple feature point information from a user for the input medical image; and An anatomical feature point automatic measurement system including a feature point measurement unit that applies the input feature points to a pre-established algorithm to generate feature point segment information and feature point angle information.

Images