Apparatus and Method for Calculating TPA for Cruciate Ligament Rupture determination of Companion Animal

Abstract

A method according to an embodiment of the present disclosure is a computer-implemented method using a TPA calculation device for determining an anterior cruciate ligament rupture in a pet, and a computer-implemented method using a TPA (Tibial Plateau Angle) calculation device for determining an anterior cruciate ligament rupture in a pet, comprising: a step of inputting images of the tibia and femur of the pet into an artificial neural network model; a step of extracting the outlines of the tibia and femur using the artificial neural network model; a step of determining a first to fourth position for measuring TPA from at least one point constituting the outlines; and a step of calculating TPA based on the determined first to fourth positions.

Description

Technology Field

[0001] The present disclosure relates to an apparatus and method for calculating TPA for determining anterior cruciate ligament rupture in companion animals. Background Technology

[0002] The following description merely provides background information related to the present embodiment and does not constitute prior art.

[0003] Figure 1 is an example of the cruciate ligament of a conventional pet animal.

[0004] The companion animals described below are, for example, dogs, but are not limited thereto.

[0005] The cruciate ligaments are ligaments that exist in the knee joint and connect the femur and the tibia. The femur is also referred to as the thigh bone (Femur) (110), and the tibia is also referred to as the shin bone (Tibia) (100).

[0006] The cruciate ligaments are mainly composed of the anterior cruciate ligament (120) and the posterior cruciate ligament (130). In addition, reference number 140 indicates the meniscus, and reference number 150 indicates the kneecap.

[0007] One of the common musculoskeletal diseases in dogs is a cranial cruciate ligament rupture (CCLR).

[0008] Figure 2 is an example diagram showing a normal kneecap and a case of anterior cruciate ligament rupture.

[0009] Figure 2(a) is an example of a normal kneecap, and Figure 2(b) is an example of an anterior cruciate ligament rupture. An anterior cruciate ligament rupture refers to a case where the ligament holding the tibia is torn, and a force is applied, causing the tibia to be pushed forward, as shown in Figure 2(b).

[0010] The anterior cruciate ligament (120) prevents the tibia (100) from moving forward or rotating inward and prevents over-relaxation of the knee joint. However, if the anterior cruciate ligament is torn, limping (e.g., abnormal gait) and pain occur.

[0011] The aforementioned anterior cruciate ligament (ACL) rupture is mostly caused progressively by chronic degenerative arthritis, and ligament damage can also occur due to external impact. Furthermore, congenital bone problems, obesity, or excessive activity accelerate the rupture. Since a torn ligament does not regenerate, symptoms associated with persistent pain appear. Therefore, an accurate diagnosis through X-ray examination is necessary for risk prevention and for precise treatment even after ligament damage occurs. For this purpose, the most widely known diagnostic method for ACL rupture is the Tibial Plateau Angle (TPA) measurement test.

[0012] The TPA test is a diagnostic method that detects abnormalities in the knee joint (e.g., a torn anterior cruciate ligament) by calculating the TPA formed by the two bones that make up a dog's knee joint: the tibia and the femur. The calculated TPA can serve as the criterion for determining an anterior cruciate ligament tear in pets.

[0013] A wide TPA indicates that there is a significant risk of the knee's cruciate ligament tearing or rupturing. Although the angle varies slightly depending on the dog breed, it ranges from approximately 22.62 to 34.93 degrees.

[0014] There are several preconditions when calculating TPA.

[0015] Figure 3 shows an example where the knee joint and ankle joint are aligned at 90 degrees.

[0016] One of the TPA requirements is that the knee and ankle joints must be aligned at 90 degrees, as shown in Fig. 3.

[0017] Figure 4 is an example of a Functional Tibial Line (FTL).

[0018] One of the TPA requirements is the FTL (Functional Tibial Line) (410).

[0019] FTL (410) is a line that can indicate the central point (400) of the talus and the central point (400) of the slightly raised intercondylar eminences.

[0020] Figure 5 is an example of the MTPL (Medial Tibial Plateau Line).

[0021] One of the TPA requirements is the MTPL (Medial Tibial Plateau Line) (520).

[0022] MTPL (520) refers to a line that can connect points (530) connecting the anterior and posterior parts of the tibia.

[0023] Figure 6 is an example diagram showing the TPA calculation method.

[0024] TPA means the angle between MTPL and the vertical line (610), as shown in FIG. 6.

[0025] Meanwhile, conventionally, veterinarians diagnosed anterior cruciate ligament rupture by directly measuring based on precisely captured images (X-rays taken with the kneecap and ankle bones at right angles) or by analyzing the captured images using artificial intelligence (AI). However, this method had the problem of being inefficient in diagnosing anterior cruciate ligament rupture due to reasons such as a lack of sample data and human intervention. A detailed explanation will be provided with reference to Fig. 7.

[0026] Figure 7 is an example diagram showing the case where the location of a joint part is found in the shape of a rectangular box using an object detection method.

[0027] Referring to FIG. 7, regions A and B identify joints, 'e' identifies the center of the talus, and 'd1' and 'd2' identify the points forming the MTPL.

[0028] Conventionally, an object detection method is used to find the position of a joint region in the form of a rectangular box as shown in FIG. 7, and to find all points required for TPA measurement at once. In other words, conventionally, an object detection method was used to find the structure of the upper end of the tibia and the structure of the lower end of the femur in the form of a rectangular box. The two found rectangular boxes serve to indicate the joint regions of the tibia and the femur.

[0029] Independently of the found boxes, the values 'd1', 'd2', and 'e' are separately obtained, without obtaining any information from the already found rectangular boxes of the tibia and femur, and the values 'd1', 'd2', and 'e' are directly located. In this case, if sufficient training data is not collected, there is a problem that it is difficult to accurately determine the positions of the TPA measurement position values 'd1', 'd2', and 'e'. In other words, the object detection method has a problem that it is difficult to find the points required for accurate TPA measurement when sufficient training data is not available. Prior Art Documents

[65535] International Academic Journal. Tonima, Masuda Akter, et al. "Auto-Detection of Tibial Plateau Angle in Canine Radiographs Using a Deep Learning Approach." 2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS). IEEE, (2021.)[Searched on September 25, 2025], Internet: <https: / / doi.org / 10.1109 / DDCLS52934.2021.<9455467> The problem to be solved

[0030] The present disclosure provides an apparatus and method for efficiently calculating TPA for determining anterior cruciate ligament rupture in companion animals.

[0031] The present disclosure provides an apparatus and method for automatically calculating the criteria for determining anterior cruciate ligament rupture in pets.

[0032] The present disclosure provides an apparatus and method for efficiently determining whether a pet has a ruptured anterior cruciate ligament.

[0033] The problems that the present invention aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0034] A method according to an embodiment of the present disclosure is a computer-implemented method using a TPA calculation device for determining anterior cruciate ligament rupture in a pet, comprising: a step of inputting images of the tibia and femur of the pet into an artificial neural network model; a step of extracting the outlines of the tibia and femur using the artificial neural network model; a step of determining a first to fourth position for TPA measurement from at least one point constituting the outlines; and a step of calculating TPA based on the determined first to fourth positions.

[0035] An apparatus according to an embodiment of the present disclosure is an apparatus for calculating TPA for determining a rupture of the cruciate ligament of a pet, comprising: a memory including instructions; and a processor that inputs an image of the tibia and femur of the pet into an artificial neural network model by executing the instructions, extracts an outline of the tibia and femur using the artificial neural network model, determines a first to fourth position for measuring TPA from at least one point constituting the outline, and calculates TPA based on the determined first to fourth position. Effects of the invention

[0036] The present disclosure can automatically calculate the criteria for determining anterior cruciate ligament rupture in pets.

[0037] The present disclosure can efficiently calculate whether a pet has a torn anterior cruciate ligament.

[0038] The present disclosure eliminates the need for a training data filtering process, allowing a much larger number of X-ray images to be used in the training data.

[0039] The present disclosure allows for the use of large X-ray images as training data by utilizing polygon learning aimed at extracting outlines.

[0040] The present disclosure enables explicit verification of error occurrence through the introduction of an outline extraction method, and thereby allows for an easy virtuous cycle structure by rapidly correcting training data and retraining it.

[0041] In this disclosure, since the knee joint and ankle joint do not necessarily need to be aligned at a 90-degree angle, all X-ray images showing the entire image can be used as training data.

[0042] The effects of the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0043] Figure 1 is an example of the cruciate ligament of a conventional pet animal. Figure 2 is an example diagram showing a normal kneecap and a case where the anterior cruciate ligament is ruptured. Figure 3 shows an example where the knee joint and ankle joint are aligned at 90 degrees. Figure 4 is an example of a Functional Tibial Line (FTL). Figure 5 is an example of the MTPL (Medial Tibial Plateau Line). Figure 6 is an example diagram showing the TPA calculation method. Figure 7 is an example diagram showing the case where the location of a joint part is found in the shape of a rectangular box using an object detection method. FIG. 8 is a first example diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 9 is a second exemplary diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 10 is a third exemplary diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 11 is a fourth example diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 12 is a block diagram of a device for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 13 is a flowchart illustrating a first method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 14 is a flowchart illustrating a second method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 15 is a flowchart illustrating a TPA calculation method according to an embodiment of the present disclosure. Specific details for implementing the invention

[0044] Some embodiments of the present disclosure are described in detail below with reference to exemplary drawings. It should be noted that in assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings. Furthermore, in describing the present disclosure, if it is determined that a detailed description of related known components or functions could obscure the essence of the present disclosure, such detailed description is omitted.

[0045] In describing the components of the embodiments according to the present disclosure, symbols such as first, second, i), ii), a), b), etc., may be used. These symbols are intended only to distinguish the components from other components, and the essence, order, or sequence of the components is not limited by the symbols. When a part in the specification is described as 'comprising' or 'having' a component, this means that, unless explicitly stated otherwise, it does not exclude other components but may include additional components.

[0046] The detailed description set forth below, together with the accompanying drawings, is intended to describe exemplary embodiments of the present disclosure and is not intended to represent the only embodiment in which the present disclosure may be practiced.

[0047] In this specification, the outlines of the tibia and femur to which the learning model is applied are composed of at least one point, and said at least one point may include a polygon. The polygon refers to a polygon, which is the smallest unit used to represent the three-dimensional shape of an object in 3D computer graphics, and polygons can be combined to represent a 3D object (e.g., the tibia and femur). In the embodiment of this disclosure, the at least one point constituting the outlines of the tibia and femur may be obtained from an artificial neural network model. A detailed description thereof will be provided using FIGS. 8 to 15.

[0048] FIG. 8 is a first example diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure.

[0049] In order to automatically calculate TPA for determining anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure, training data is constructed using images of the tibia and femur of the pet (e.g., images taken by X-ray), and an artificial neural network model is applied to the constructed training data to extract the outlines of the tibia and femur, and TPA can be calculated based on the polygons constituting the outlines. At this time, the measurement points (TPA measurement points) required to calculate TPA may include e, f, d1, d2, etc. The method for determining e, f, d1, and d2 will be described in detail below.

[0050] Below, we will explain in detail how to locate TPA measurement points.

[0051] In the embodiments of the present disclosure, instead of learning six points (e.g., a, b, c, e, d2, d1) required for TPA calculation based on a knee joint X-ray image in which the pet's bones are placed in a perfect alignment, a method may be used to learn the outline image of the pet's bones (e.g., tibia, femur) to find the accurate outline, and then learn the key points for TPA calculation located on the outline of the tibia. Accordingly, in the embodiments of the present disclosure, the knee joint and the ankle joint do not necessarily need to be aligned at a 90-degree angle. Furthermore, in the embodiments of the present disclosure, all X-ray images including the entire image of the tibia can be used as training data.

[0052] Referring to FIG. 8, a method for determining the location of 'e', ​​one of four measurement points according to an embodiment of the present disclosure, for TPA measurement will be described.

[0053] The point at reference number 820 is a point representing the outline of the tibia, and the point at reference number 830 is a point representing the outline of the femur.

[0054] The position of 'e' can be determined as the midpoint (or midpoint) of the two closest points among point 820 (reference number 820) and point 830 (reference number 830) inside the circle (810) of FIG. 8. In other words, the position of 'e' can be determined as the midpoint of the coordinates of the two closest points of the two outlines of the tibia and femur.

[0055] FIG. 9 is a second exemplary diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure.

[0056] In the process of finding measurement points 'd1', 'd2', and 'e', ​​instead of calculating all the numerous polygons found to accurately represent the outlines of the tibia and femur, the amount of calculation can be reduced by drawing a circle around the area with the highest polygon density (e.g., the area where the tibia and femur meet) and then finding 'd1', 'd2', and 'e' from the polygons inside that circle.

[0057] At this time, the method for determining the position and size of the circle will be explained in Fig. 9, which will be described later. The method for determining 'd1' and 'd2' will be explained in Fig. 11, which will be described later.

[0058] The method of drawing a circle involves finding all (x, y) for all polygon coordinates of the tibia, finding max(x) representing the largest value on the x-axis, max(y) representing the largest value on the y-axis, min(x) representing the smallest value on the x-axis, and min(y) representing the smallest value on the y-axis, and then subtracting min(y) from max(y) to calculate the approximate length of the tibia.

[0059] A circle (920) with radius r is drawn at ¼ of the approximate length of the calculated tibia, wherein the x-coordinate value of the center of the circle (920) is determined as the midpoint between max(x) and min(x), and the y-coordinate value of the center of the circle (920) can be determined as max(y).

[0060] Referring to FIG. 9, the straight line associated with the circle (920) may include a straight line of the approximate length of the tibia calculated by subtracting min(y) from max(y), and a straight line of the length of a segment that is ¼ of the approximate length of the tibia.

[0061] FIG. 10 is a third exemplary diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure.

[0062] Referring to Fig. 10, we will explain how to determine the location of 'f', which is the middle part of the lower tibia.

[0063] Search for the point with the smallest y-coordinate value among all polygons of the tibia. The point with the smallest y-coordinate value among all polygons of the tibia can be any location in the lower part of the tibia.

[0064] Starting from the point with the smallest y-coordinate value among all the polygon coordinates of the tibia, you can search for the next coordinates by moving all the coordinates forming the polygon in both left and right directions. Alternatively, you can search for other coordinates by moving up and down directions centered on the point with the smallest y-coordinate value.

[0065] When comparing the values ​​of consecutive coordinates searched in the direction of arrow A (1020) and arrow B (1010) in FIG. 10, one of the directions forms an arch shape (1030). The arch shape is the joint area of ​​the lower part of the tibia. The median value of the coordinate values ​​forming this joint area can be found, and the location of the found median value can be determined as the measurement point 'f' of the lower part of the tibia. That is, the median value of the consecutive coordinates of the arch shape can be determined as the measurement point 'f'.

[0066] FIG. 11 is a fourth example diagram illustrating a method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure.

[0067] FIG. 11 is an example diagram of a method for determining the values ​​of measurement points 'd1' and 'd2' for TPA measurement according to an embodiment of the present disclosure.

[0068] Conventionally, the method involved using object detection to locate rectangles around the joint area to roughly determine the positions of the upper part of the tibia and the lower part of the femur, thereby finding measurement points. However, this method has a problem in that the bounding box of the rectangle does not exactly align with the measurement points. Consequently, the measurement points are not always located on the outer edge of the rectangle.

[0069] However, the embodiment of the present disclosure focuses on finding the outline of the bone in the form of a polygon to find the accurate outline of the bone, and since the TPA measurement points 'd1' and 'd2' are always found on the polygon forming the outline of the tibia, the joint area does not need to be right angle. The positions of 'd1' and 'd2' according to the embodiment of the present disclosure can be determined using outline information (e.g., polygon position information) rather than by obtaining possible positions through neural network learning.

[0070] A method for determining the positions of d1' and 'd2' according to an embodiment of the present disclosure is as follows.

[0071] First, draw a straight line connecting the predetermined 'e' and 'f'.

[0072] In reality, when the coordinates of 'e' and 'f' are calculated and connected by a straight line, a vertical line is not always formed. Therefore, the values ​​of 'e' and 'f' may be determined depending on the angle of the captured X-ray image. At this time, as shown in Fig. 11, the straight line connecting 'e' and 'f' is rotated so that it becomes a vertical line on the screen, making it easier to calculate the remaining coordinate values. In other words, the straight line is rotated so that the x-coordinate values ​​of 'e' and 'f' become equal.

[0073] A line perpendicular to the above straight line is used to find the point where it intersects the outline of the tibia at coordinate 'e' while gradually increasing the y-axis value. The coordinate on the outline of the intersection point, where the y-axis value is highest, can be the location 'd2' (indicated as ① in Fig. 11).

[0074] In the same way, a line perpendicular to the above line is positioned to intersect the outline of the tibia by gradually lowering the y-axis value at coordinate 'e'. The coordinate on the outline where it intersects when the x-axis value is greatest can be the position of 'd1' (indicated as ② in Fig. 11). When determining the position of d2', parts deviating from the section described in Fig. 9 may not be calculated.

[0075] The positions of 'd1' and 'd2' determined in this way are located on the outline of the tibia.

[0076] FIG. 12 is a block diagram of a device for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure.

[0077] Referring to FIG. 12, the apparatus for calculating TPA for determining a cruciate ligament rupture in a pet according to an embodiment of the present disclosure may be configured to include a learning data configuration unit (1201), a CNN (Convolutional Neural Network), an outline extraction unit (1203), a TPA calculation unit (1204), a comparison unit (1205), a learning data error correction unit (1206), etc.

[0078] The learning data composition unit (1201) can compose the learning data using images of the tibia and femur of a pet. The captured images may include, for example, images taken with an X-ray, computed tomography (CT) images, magnetic resonance imaging (MRI), positron emission tomography (PET) images, etc.

[0079] The artificial neural network model unit (1202) can apply an artificial neural network model to training data. The artificial neural network model may include, for example, a CNN model. The CNN model may use a basic CNN structure and an improved CNN structure. The CNN model may use, for example, CNN, R-CNN, Fast R-CNN (Fast Region-based Convolutional Neural Network), Faster R-CNN, and Mask R-CNN.

[0080] The outline extraction unit (1203) can extract the outlines of the tibia and femur as a result of applying an artificial neural network model to the training data. Additionally, the outline extraction unit (1203) can extract the inference results of the tibia and femur as a result of applying an artificial neural network model to the training data.

[0081] The TPA calculation unit (1204) can determine a TPA (Tibial Plateau Angle) measurement point from at least one point included in the outline and calculate the TPA based on the TPA measurement point. The at least one point may include a polygon. At this time, the four required coordinate values ​​(TPA measurement points) may be the positions 'd1', 'd2', 'e', ​​and 'f'.

[0082] The comparison unit (1205) can compare the points of the outline output from the outline extraction unit (1203) with actual measurement data (ground truth). The actual measurement data can be referred to as actual measurement data.

[0083] The training data error correction unit (1206), when the comparison result of the comparison unit (1205) indicates that the points of the outline output from the outline extraction unit (1203) do not match the actual measurement data, corrects the training data and then transmits it back to the training data configuration unit (1201). The corrected training data can be fed back into the training data.

[0084] The operation of the components of FIG. 12 (training data composition unit (1201), CNN (Convolutional Neural Network), outline extraction unit (1203), TPA calculation unit (1204), comparison unit (1205), training data error correction unit (1206)) can be performed in a processor. That is, a device for calculating TPA for determining a cruciate ligament rupture in a pet may include a memory containing instructions; and a processor that inputs an image of the tibia and femur of the pet into an artificial neural network model by executing the instructions, extracts the outlines of the tibia and femur using the artificial neural network model, determines a TPA measurement point from at least one point constituting the outline, and calculates TPA based on the TPA measurement point.

[0085] FIG. 13 is a flowchart illustrating a first method for calculating TPA for determining an anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure.

[0086] The learning data composition unit (1201) can compose the learning data with an image taken with an X-ray in step 1301.

[0087] The artificial neural network model section (1202) can apply the artificial neural network model to the training data in step 1302 and output it.

[0088] The outline extraction unit (1203) can extract the outlines of the tibia and femur based on the output of the artificial neural network model unit (1202) in step 1303.

[0089] The TPA calculation unit (1204) can calculate TPA from the outline extracted in step 1304 and the outline (polygon) of the femur.

[0090] FIG. 14 is a flowchart illustrating a second method for calculating TPA for determining anterior cruciate ligament rupture in a pet according to an embodiment of the present disclosure. FIG. 14 illustrates the process of calculating TPA by providing the outline extraction inference result as an input to the TPA calculation unit.

[0091] The learning data composition unit (1201) can compose the learning data with an image taken with an X-ray in step 1401.

[0092] The artificial neural network model unit (1202) can apply the artificial neural network model to the training data in step 1402 and output it. The artificial neural network model may include, for example, a CNN model, and the CNN model may use a basic CNN structure and an improved CNN structure.

[0093] The outline extraction unit (1203) can extract the outlines of the tibia and femur based on the output of the artificial neural network model unit (1202) in step 1403. The outlines include at least one point, and the at least one point may include a polygon.

[0094] The comparison unit (1205) can compare the inference results of the model created through learning in step 1404 with actual data.

[0095] If the inference result of the model created through learning does not match the actual data, the learning data error correction unit (1206) can correct the learning data in step 1405 and then send it back to the learning data composition unit (1201) to be input into the learning data.

[0096] FIG. 15 is a flowchart illustrating a TPA calculation method according to an embodiment of the present disclosure.

[0097] The TPA calculation unit (1204) can determine the location of the TPA measurement point 'e' in step 1501. Since the method for determining the location of 'e' has been explained with reference to FIG. 8, a detailed explanation will be omitted.

[0098] The TPA calculation unit (1204) can determine the location of the TPA measurement point 'f' in step 1502. Since the method for determining the location of 'f' has been explained with reference to FIG. 10, a detailed explanation will be omitted.

[0099] The TPA calculation unit (1204) can generate a first straight line connecting TPA measurement points 'e' and 'f' in step 1503.

[0100] The TPA calculation unit (1204) can rotate the first straight line generated to match the x-axis coordinate values ​​in step 1504.

[0101] The TPA calculation unit (1204) can determine the location of the TPA measurement point 'd2' in step 1505. A line perpendicular to the first straight line is used to find the location where it intersects the outline of the tibia by gradually increasing the y-axis value at the 'e' coordinate. The coordinates on the outline of the point where it intersects when the y-axis value is at its maximum can be the location of 'd2'.

[0102] The TPA calculation unit (1204) can determine the location of the TPA measurement point 'd1' in step 1506. A line perpendicular to the first straight line is used to find the location where it intersects the outline of the tibia by gradually lowering the y-axis value at the 'e' coordinate. The coordinate on the outline where it intersects when the x-axis value is greatest can be the location of 'd1'.

[0103] The method for determining the positions of 'd2' and 'd1' has been explained with reference to Fig. 11, so a detailed explanation will be omitted.

[0104] The TPA calculation unit (1204) can generate a second straight line connecting TPA measurement points 'd1' and 'd2' in step 1507.

[0105] The TPA calculation unit (1204) can calculate TPA at the intersection of the first straight line and the second straight line in step 1508.

[0106] At least some of the components described in the exemplary embodiments of the present disclosure may be implemented as hardware elements including at least one of a Digital Signal Processor (DSP), a processor, a network controller, an Application-Specific IC (ASIC), a programmable logic device (such as an FPGA), or other electronic devices, or a combination thereof. Additionally, at least some of the functions or processes described in the exemplary embodiments may be implemented in software, and the software may be stored on a recording medium. At least some of the components, functions, and processes described in the exemplary embodiments of the present disclosure may be implemented as a combination of hardware and software.

[0107] The method according to exemplary embodiments of the present disclosure may be written as a program executable on a computer and may also be implemented on various recording media such as magnetic storage media, optical reading media, and digital storage media.

[0108] Implementations of the various technologies described herein may be implemented as digital electronic circuits, or as computer hardware, firmware, software, or combinations thereof. Implementations may be implemented as computer program products, i.e., computer programs tangibly embodied in information carriers, such as machine-readable storage devices (computer-readable media) or radio signals, for processing by the operation of data processing devices, e.g., programmable processors, computers, or multiple computers, or for controlling such operation. Computer programs such as the computer program(s) described above may be written in any form of programming language, including compiled or interpreted languages, and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. Computer programs may be deployed to be processed on one computer or multiple computers at one site, or distributed across multiple sites and interconnected by a communication network.

[0109] Processors suitable for processing computer programs include, for example, both general-purpose and special-purpose microprocessors, and any one or more processors of any type of digital computer. Generally, the processor will receive instructions and data from read-only memory or random access memory, or both. The elements of the computer may include at least one processor that executes instructions and one or more memory devices that store instructions and data. Generally, the computer may include one or more mass storage devices that store data, for example, magnetic, magneto-optical disks, or optical disks, or may be combined to receive data from these, transmit data to these, or both. Information carriers suitable for embodying computer program instructions and data include, for example, semiconductor memory devices, magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROMs (Compact Disk Read Only Memory) and DVDs (Digital Video Disks); magneto-optical media such as floptical disks; ROMs (Read Only Memory); RAMs (Random Access Memory); flash memory; EPROMs (Erasable Programmable ROM); EEPROMs (Electrically Erasable Programmable ROM); etc. Processors and memory may be supplemented by or included in special-purpose logic circuit organizations.

[0110] The processor can execute an operating system and software applications executed on said operating system. Additionally, the processor device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, the processor device may be described as being used as a single unit; however, those skilled in the art will understand that the processor device may include multiple processing elements and / or multiple types of processing elements. For example, the processor device may include multiple processors or one processor and one network controller. Other processing configurations, such as a parallel processor, are also possible.

[0111] Additionally, non-transitory computer-readable media may be any available medium accessible by a computer and may include both computer storage media and transmission media.

[0112] Although this specification contains details of a number of specific embodiments, they should not be understood as limiting the scope of any invention or claimables, but rather as descriptions of features that may be characteristic of a specific embodiment of a specific invention. Specific features described in this specification in the context of individual embodiments may be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented in multiple embodiments individually or in any appropriate sub-combination. Furthermore, while features may operate in a specific combination and be described as initially claimed, one or more features from the claimed combination may be excluded from the combination in some cases, and the claimed combination may be changed to a sub-combination or a variation of the sub-combination.

[0113] Likewise, although operations are depicted in the drawings in a specific order, this should not be understood as requiring that such operations be performed in that specific or sequential order depicted to obtain a desirable result, or that all depicted operations must be performed. In certain cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of the various device components of the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and devices can generally be integrated together into a single software product or packaged into multiple software products.

[0114] Meanwhile, the embodiments disclosed in this specification and drawings are merely specific examples provided to aid understanding and are not intended to limit the scope of the invention. It is obvious to those skilled in the art that other variations based on the technical concept of the invention are possible in addition to the embodiments disclosed herein.

[0115] The scope of protection of this embodiment shall be interpreted by the claims below, and all technical ideas within an equivalent scope shall be interpreted as being included within the scope of rights of this embodiment.

Claims

Claim 1 A method performed by a TPA (Tibial Plateau Angle) calculation device for determining an anterior cruciate ligament rupture in a pet, comprising: a process of inputting images of the tibia and femur of the pet into an artificial neural network model; a process of extracting the outlines of the tibia and femur using the artificial neural network model; and a process of determining first to fourth positions for TPA measurement from at least one point constituting the outlines. A method comprising a process of calculating TPA based on the first to fourth positions determined above, wherein the first position is determined based on the minimum distance position between the outline of the tibia and the outline of the femur, the second position is determined based on the total length of coordinates arranged continuously according to the distal tibial region located opposite to the first position among the outlines of the tibia, and the third and fourth positions are determined as poles of the outline of the tibia in a direction perpendicular to the first straight line in a coordinate system rotated with respect to the first straight line connecting the first position and the second position, wherein the third position is a pole in the perpendicular direction and the fourth position is a pole in a direction orthogonal to the perpendicular direction. Claim 2 A method according to claim 1, wherein the determining process comprises: a process of generating a first straight line using the first position and the second position; and a process of generating a second straight line using the third position and the fourth position. Claim 3 A method according to claim 1, wherein the process of determining the first position comprises determining the first position as the midpoint of two points of the closest distance between the tibia and the femur from at least one point constituting the outline. Claim 4 A method according to claim 1, wherein the process of determining the second position comprises: a process of searching for coordinates while moving left and right around the point with the smallest y-axis coordinate value from at least one point constituting the lower end of the tibia; and a process of determining the median value of the total length of consecutive coordinates representing the joint portion of the lower end of the tibia among the searched coordinates as the second position. Claim 5 A method according to claim 2, wherein the process of determining the third position comprises: a process of rotating the first straight line so that the x-axis coordinate values ​​of the first position and the second position become equal; and a process of determining the third position as the position where the y-axis value is maximum among the positions where a line perpendicular to the rotated first straight line meets the outline of the tibia. Claim 6 A method according to claim 2, wherein the process of determining the fourth position comprises: a process of rotating the first straight line so that the x-axis coordinate values ​​of the first position and the second position become equal; and a process of determining the fourth position as the position having the maximum x-axis value among the positions where a line perpendicular to the rotated first straight line and the outline of the tibia intersect. Claim 7 The method according to claim 1, wherein the process of determining the third and fourth positions comprises: a process of extracting a maximum value of the x-axis, a minimum value of the x-axis, a maximum value of the y-axis, and a minimum value of the y-axis in two dimensions from at least one point constituting the outline of the tibia; a process of calculating the total length of the tibia by subtracting the minimum value of the y-axis from the maximum value of the y-axis; a process of generating a circle based on a preset ratio value for the total length of the tibia; and a process of determining the third and fourth positions from at least one point included within the generated circle. Claim 8 A method according to claim 7, wherein the x-axis coordinate of the center point of the circle is determined as the midpoint between the maximum value of the x-axis and the minimum value of the x-axis, and the y-axis coordinate of the center point of the circle is determined as the maximum value of the y-axis. Claim 9 In paragraph 2, the process of calculating the TPA is a method of calculating using the angle formed by the first straight line and the second straight line. Claim 10 A computer-readable recording medium having stored instructions, wherein the instructions, when executed by the computer, cause the computer to execute each process included in the method according to any one of claims 1 to 9. Claim 11 A device for calculating TPA for determining anterior cruciate ligament rupture in a pet, comprising: a memory containing instructions; A device comprising a processor that inputs an image of the tibia and femur of the pet, taken by executing the above command, into an artificial neural network model, extracts the outlines of the tibia and femur using the artificial neural network model, determines a first to fourth position for TPA measurement from at least one point constituting the outline, and calculates TPA based on the determined first to fourth position, wherein the first position is determined based on the minimum distance position between the outline of the tibia and the outline of the femur, the second position is determined based on the total length of coordinates arranged continuously according to a tibial end region located opposite to the first position among the outlines of the tibia, and the third and fourth positions are determined as poles of the outline of the tibia in a direction perpendicular to the first straight line in a coordinate system rotated with respect to a first straight line connecting the first position and the second position, wherein the third position is a pole in the perpendicular direction and the fourth position is a pole in a direction orthogonal to the perpendicular direction.

Images