Method, computing apparatus for contour correction of regions in MRI images

Abstract

The present invention relates to a technique for correcting the contours of a structure within an MRI image, and may include the steps of generating a mask image for the contours of a structure within an MRI image, generating a skeleton for the contours within the generated mask image, modifying the mask image through image processing using the skeleton, and correcting the contours of a structure within an MRI image through image processing using the modified mask image. According to the present invention, by correcting disconnected contours within an MRI image, it is possible to identify features of the structure.

Description

Technology Field

[0001] The present invention relates to a technique for correcting the contours of structures within an MRI image. Background Technology

[0002] MRI (Magnetic Resonance Imaging) is a medical imaging technology that uses strong magnetic fields and radio waves to obtain images of various organs and structures of the human body.

[0003] MRI images obtained from an MRI device provide information on various interpretation items, and users utilize these images for disease diagnosis, interpretation of trauma, establishment of treatment plans, and monitoring of progress.

[0004] Specifically, MRI images provide high-resolution, detailed visualizations that allow for the clear identification of structural abnormalities in organs or tissues, tumors, and inflammatory conditions. This enables users to make accurate diagnoses and design personalized treatment plans, while also playing a crucial role in tracking post-treatment progress and evaluating effectiveness.

[0005] In addition, to provide users with more accurate MRI images and to display only the organs subject to judgment, users utilize artificial neural network models to identify structures within the MRI images or correct the images through image processing.

[0006] For example, a conventional method using an artificial neural network model can identify structures, such as organs or tissues, that the user intends to evaluate in an MRI image, and generate contours of the structures using the artificial neural network model to enhance the identification accuracy of the structures to be evaluated.

[0007] However, conventional methods result in the thinnest parts of the contours generated through artificial neural network models being discontinuously disconnected.

[0008] This makes it difficult to distinguish the precise boundaries of the structure during subsequent image processing of the MRI images, which can lower the reliability of diagnosis based on MRI images.

[0009] Therefore, in order to improve the quality and increase the reliability of MRI images, a method is required to correct disconnected contours generated through artificial neural network models in MRI images. The problem to be solved

[0010] The present invention aims to provide a method for contour correction of structures within an MRI image and a computing device for performing the same.

[0011] This specification is not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0012] A method for correcting the contours of a structure within an MRI image, performed on a computing device according to an embodiment of the present invention for achieving the above-described purpose, may include the steps of generating a mask image for the contours of a structure within an MRI image, generating a skeleton for the contours within the generated mask image, modifying the mask image through image processing using the skeleton, and correcting the contours of the structure within the MRI image through image processing using the modified mask image.

[0013] In addition, the step of generating the skeleton may generate the skeleton by expanding the contours within the mask image and connecting the inner centers of the expanded contours.

[0014] In addition, the above-mentioned modification step can modify disconnected contours within the mask image through image processing that composites the skeleton onto the mask image.

[0015] In addition, the above correction step can correct disconnected contours within the MRI image through image processing that synthesizes a modified mask image onto the MRI image.

[0016] In addition, it may further include a step of generating contours for structures within the MRI image using an artificial neural network model.

[0017] Additionally, the method further includes a step of determining whether the generated contour is disconnected, and the step of generating the mask image can generate a mask image for the disconnected contour.

[0018] In addition, the above-mentioned decision-making step may be determined based on the number of regions of interest adjacent to a specific part of the contour line.

[0019] Meanwhile, the computing device includes a processor, a memory for loading a computer program executed by the processor, and a storage for storing the computer program, wherein the computer program may include an operation to generate a mask image for the contours of a structure within an MRI image, an operation to generate a skeleton for the contours within the generated mask image, an operation to modify the mask image through image processing using the skeleton, and an operation to correct the contours of the structure within the MRI image through image processing using the modified mask image.

[0020] In addition, the operation of generating the skeleton can generate the skeleton by expanding the contours within the mask image and connecting the inner centers of the expanded contours.

[0021] In addition, the above modification operation can modify disconnected contours within the mask image through image processing that composites the skeleton onto the mask image.

[0022] In addition, the above-mentioned correction operation can correct disconnected contours within the MRI image through image processing that synthesizes a modified mask image onto the MRI image.

[0023] In addition, it may further include an operation to generate contours of structures within the MRI image using an artificial neural network model.

[0024] Additionally, the operation of determining whether the generated contour is disconnected is further included, and the operation of generating the mask image can generate a mask image for the disconnected contour.

[0025] In addition, the above-mentioned judgment operation can be determined based on the number of regions of interest adjacent to a specific part based on a specific part of the contour.

[0026] Meanwhile, a computer-readable recording medium according to one embodiment of the present invention for achieving the above-described purpose may store a program for performing the above-described contour correction method.

[0027] In addition, a computer program stored on a computer-readable recording medium according to one embodiment of the present invention for achieving the above-described purpose may include program code for executing the above-described contour correction method. Effects of the invention

[0028] According to the present specification, the present invention has the effect of identifying structural features by correcting disconnected contours within an MRI image.

[0029] In addition, the present invention has the effect of enabling more accurate image segmentation after post-processing of MRI images by correcting disconnected contours within MRI images.

[0030] The effects of the present invention 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

[0031] FIG. 1 is a conceptual diagram showing a contour correction system for a structure within an MRI image according to one embodiment of the present invention. FIG. 2 is a flowchart illustrating a contour correction method according to one embodiment of the present invention. FIG. 3 is an exemplary diagram showing a contour line within an MRI image according to one embodiment of the present invention. FIG. 4 is an exemplary diagram showing a segmented contour within an MRI image according to one embodiment of the present invention. FIG. 5 is an exemplary diagram showing a post-processed MRI image according to one embodiment of the present invention. FIG. 6 is a flowchart illustrating a contour correction method according to one embodiment of the present invention in more detail. FIG. 7 is an exemplary diagram illustrating the process of modifying contours within a mask image according to one embodiment of the present invention. FIG. 8 is an exemplary diagram illustrating a process for generating a corrected MRI image according to one embodiment of the present invention. FIG. 9 is an exemplary diagram illustrating a process of generating a corrected MRI image based on the position coordinates of contours within an MRI image and contours within a modified mask image according to an embodiment of the present invention. FIG. 10 is an exemplary diagram showing the configuration of a computing device that performs a contour correction method according to one embodiment of the present invention. Specific details for implementing the invention

[0032] The following description merely illustrates the principles of the present invention. Therefore, those skilled in the art may invent various devices that embody the principles of the present invention and are included within the concept and scope of the present invention, even though they are not explicitly described or illustrated in this specification.

[0033] Furthermore, all conditional terms and embodiments listed in this specification are, in principle, explicitly intended only for the purpose of enabling an understanding of the concept of the invention and should be understood not as being limited to the embodiments and conditions specifically listed as such.

[0034] The aforementioned objectives, features, and advantages will become clearer through the following detailed description in conjunction with the attached drawings, and accordingly, a person skilled in the art to which the present invention pertains will be able to easily implement the technical concept of the present invention.

[0035] In addition, in describing the present invention, if it is determined that a detailed description of known technology related to the present invention may unnecessarily obscure the essence of the present invention, such detailed description will be omitted.

[0036] Various embodiments of the present invention will be described in detail below with reference to the attached drawings.

[0037] FIG. 1 is a conceptual diagram showing a contour correction system for a structure within an MRI image according to one embodiment of the present invention.

[0038] Referring to FIG. 1, the MRI device (10) can generate an MRI image from a subject and transmit the generated MRI image to a computing device (300).

[0039] Here, the MRI device (10) is a magnetic resonance imaging device, meaning a device that non-invasively acquires a diagnostic image of a subject using a magnetic field generated by magnetic force.

[0040] In addition, while the subjects of examination can be exemplified by the human body, this is not limited to this and may also apply to other animals.

[0041] In addition, MRI images can represent internal organs of the human body (e.g., brain, heart, liver, kidney, and pancreas, etc.), and in the present invention, a structure may refer to an internal organ or tissue within the human body within the MRI image.

[0042] And, the computing device (300) can generate a contour of a structure within an MRI image and, if a break occurs in the generated contour, correct the broken contour to provide an MRI image.

[0043] Accordingly, the computing device (300) can provide a corrected MRI image to the user, thereby providing convenience in MRI image interpretation.

[0045] Hereinafter, a method for correcting the contour of a computing device (300) will be described in detail with reference to FIG. 2.

[0046] FIG. 2 is a flowchart illustrating a contour correction method according to one embodiment of the present invention.

[0047] Referring to FIG. 2, the computing device (300) can acquire an MRI image from an MRI device (S100).

[0048] Specifically, the computing device (300) can acquire an MRI image of the subject being examined generated from an MRI device.

[0049] Meanwhile, embodiments of the present invention are described with respect to the heart and MRI images of the heart, but are not limited thereto.

[0050] Here, the MRI image may correspond to any of the cardiac magnetic resonance imaging (Cardiac MRI) of the heart, and among them, the Cardiac CINE 4-chamber view image used to evaluate the structure and function of the heart may correspond to this.

[0051] Specifically, apical quadrilateral imaging can be used to determine the dimensions of the four parts of the heart and to observe and evaluate the structure of each part of the heart and blood flow in real time.

[0052] Hereinafter, MRI images in this specification will be described based on the apical four-cardiac view.

[0054] Next, the computing device (300) can generate a contour of a structure within an MRI image of the acquired heart (S200).

[0055] Here, the contour is a line that further emphasizes the boundaries of each structure within the MRI image, and the corresponding structure may be colored in a unique color or output as a boundary line so that each structure can be distinguished from one another.

[0056] Specifically, the computing device (300) can use an artificial neural network model to distinguish structures within an MRI image and generate outlines for the distinguished structures.

[0057] Here, the artificial neural network model is trained to distinguish structures of the heart in MRI images of the heart, and the artificial neural network model may be an image processing model such as a CNN (Convolutional Neural Networks) composed of layers that perform multiple convolution operations.

[0058] Meanwhile, the computing device (300) may generate a contour line for only a specific structure of the organ, or generate a contour line for all structures constituting the organ. For example, it may generate a contour line for all structures of the heart (left ventricle, left atrium, right ventricle and right atrium), or generate a contour line for only the left ventricle.

[0059] In this regard, further explanation is provided with reference to Fig. 3.

[0060] FIG. 3 is an exemplary diagram showing a contour line within an MRI image according to one embodiment of the present invention.

[0061] Referring to FIG. 3, the computing device (300) can use an artificial neural network model to generate contours representing the boundaries of structures of the heart within an MRI image. At this time, the computing device (300) can display contours representing the boundaries of the left atrium, left ventricle, right atrium, and right ventricle constituting the heart in different colors, respectively.

[0062] For example, the computing device (300) may display the outline of the left atrium (31) in blue, the outline of the left ventricle (32) in red, the outline of the right atrium (33) in yellow, and the outline of the right ventricle (34) in green.

[0063] Additionally, the computing device (300) may display the disconnected contour in a different color from the undisconnected contour depending on whether the disconnected contour is determined later.

[0065] Meanwhile, the computing device (300) can generate a contour line in which a specific part is disconnected due to the process of generating a contour line using an artificial neural network model or due to issues such as the resolution of the MRI image.

[0066] Here, specific parts may refer to areas where the boundaries between structures within the MRI image are ambiguous, or areas where the structure is thin or angular due to the characteristics of the structure.

[0067] Furthermore, the discontinuity does not refer to an actual structural break, but rather to a break in the contour that occurs in a specific part of the contour during the contour generation process due to structural characteristics.

[0068] For example, a discontinuous contour in a specific part of the heart can be found in the apex, a pointed and thin structure within the heart.

[0069] In this regard, this will be explained in detail with reference to FIGS. 4 and FIGS. 5.

[0070] FIG. 4 is an exemplary diagram showing a segmented contour within an MRI image according to one embodiment of the present invention.

[0071] FIG. 5 is an exemplary diagram showing a post-processed MRI image according to one embodiment of the present invention.

[0072] Referring to FIG. 4, the computing device (300) can generate a contour (42) corresponding to the left ventricle (41) of the heart.

[0073] At this time, a break in the contour may occur at the apex portion (43) of the left ventricle.

[0074] In addition, MRI images may be distorted during post-processing due to the discontinuity of contours.

[0075] For example, as shown in FIG. 5, based on the severed part of the contour, the contour exists on one side (51) but the contour can be removed on the opposite side (52).

[0076] Accordingly, the computing device (300) can determine a disconnected part within the contour and correct the disconnected contour (S300).

[0077] Here, correction may refer to the process of creating a continuous contour line without visually disconnected parts of the contour line.

[0078] In this regard, a more detailed explanation will be provided with reference to Fig. 6.

[0079] FIG. 6 is a flowchart illustrating a contour correction method according to one embodiment of the present invention.

[0080] Referring to FIG. 6, the computing device (300) can determine whether the generated contour is disconnected (S310).

[0081] Specifically, the computing device (300) can select a region of interest (ROI) based on pixel values ​​(color) for a specific part of the contour where disconnection is likely to occur.

[0082] In this case, if a break in the contour occurs at a specific part, there may be a broken contour and an exposed space between the broken contours.

[0083] Therefore, if no discontinuity occurs in a specific part of the contour, a single region of interest for the contour pixel values ​​can be created because only the contour exists in that specific part.

[0084] In addition, if a break occurs in a specific part of the contour, two regions of interest may be generated at that specific part: the broken contour and the space between the broken contours.

[0085] Accordingly, the computing device (300) can determine that the contour is disconnected in a specific part if at least two or more regions of interest are generated in a specific part.

[0086] That is, the computing device (300) determines whether there is a break within the contour, thereby avoiding unnecessary repeated correction work for all contours and selectively performing the correction process only when a break occurs.

[0087] Therefore, by reducing the processing time of the contour correction system and saving computational resources to increase overall processing efficiency, the performance of the correction system can be optimized even for complex contours.

[0089] And, the computing device (300) can generate a mask image for the disconnected contour (S320).

[0090] Here, the mask image is an image generated by an image processing technique to emphasize or isolate specific structures, which prevents distortion of other structures within the MRI image during contour correction and allows for the correction of only the contours.

[0091] Specifically, the computing device (300) can select an outline of interest in the MRI image.

[0092] And, the computing device (300) can generate a mask image by removing parts excluding the object of interest. That is, the mask image may be an image in which only the contour of the object of interest exists and parts that are not the object of interest are removed.

[0094] And, the computing device (300) can generate a skeleton for the contours within the mask image (S330).

[0095] Here, the skeleton is a simplified representation of the contours within the mask image and may be the centerline of the contours thinly reduced along the central axis of the contours.

[0096] Specifically, the computing device (300) can dilate the contours within the masking image to generate a skeleton.

[0097] Here, dilation can be a method of expanding the white areas of an image using a matrix (e.g., a kernel or a structuring element).

[0098] That is, the computing device (300) can expand the contour to obtain a larger contour compared to the existing contour, and due to the contour expansion, the disconnected parts of the contour can come into contact with each other. Therefore, the expanded contour may not have any disconnected parts.

[0099] And, the computing device (300) can generate a skeleton by connecting the inner centers of the expanded contours.

[0101] And, the computing device (300) can modify the mask image through image processing that synthesizes the mask image based on the generated skeleton (S340).

[0102] Specifically, the computing device (300) can superimpose the contours within the existing mask image and the skeleton through image processing that synthesizes the generated skeleton and the existing mask image.

[0103] At this time, the computing device (300) can superimpose the skeleton at the same location by matching the location coordinates of the contour within the existing mask image and the location coordinates of the starting point, specific part, and end point of the skeleton.

[0104] Accordingly, the computing device (300) can modify the contour by overlapping the contour and the skeleton within the existing mask image and filling in the disconnected parts of the contour with the skeleton.

[0105] Hereinafter, steps S320 to S340 will be explained in more detail with reference to FIG. 7.

[0106] FIG. 7 is an exemplary diagram illustrating the process of modifying contours within a mask image according to one embodiment of the present invention.

[0107] Referring to FIG. 7, the computing device (300) can select an outline (711) of interest in the MRI image (71) to correct the MRI image outline.

[0108] And, the computing device (300) can generate a mask image (72) by removing the part (712) excluding the object of interest.

[0109] That is, the mask image (72) may be an image in which only the contour (721) of interest exists and the part (722) of non-interest is removed.

[0110] And, the computing device (300) can generate a skeleton for the contours within the mask image.

[0111] Specifically, the computing device (300) can apply a kernel centered on a white pixel of a contour within a mask image to convert a neighboring area of ​​a white pixel within the contour to white, thereby expanding the contour to generate an expanded contour (73), and can generate a skeleton (74) by connecting the inner centers of the expanded contour.

[0112] And, the computing device (300) can generate a modified contour (751) by overlapping the position coordinates of the generated skeleton with the position coordinates of the start point, specific part, and end point of the contour (721) within the mask image (72).

[0113] That is, the computing device (300) can generate a modified mask image (75) through image processing that synthesizes the skeleton and mask images.

[0115] And, the computing device (300) can correct the contours of the structures within the MRI image using the modified mask image (S350).

[0116] Specifically, the computing device (300) can correct contours within the MRI image through image processing that synthesizes the modified mask image and the MRI image.

[0117] In this regard, a more detailed explanation will be provided with reference to Fig. 8.

[0118] FIG. 8 is an exemplary diagram illustrating a process for generating a corrected MRI image according to one embodiment of the present invention.

[0119] Referring to FIG. 8, the computing device (300) can compare the position coordinates of the contour (811) in the MRI image and the contour (821) in the modified mask image in order to synthesize the MRI image (81) and the modified mask image (82).

[0120] And, the computing device (300) can generate a corrected contour (831) and a corrected MRI image (83) based on the location coordinates of the contour (811) in the MRI image and the contour (821) in the modified mask image, thereby preventing the contour from being removed or distorted after image processing.

[0121] In this regard, further explanation is provided with reference to Fig. 9.

[0122] FIG. 9 is an exemplary diagram illustrating a process of generating a corrected MRI image based on the position coordinates of contours within an MRI image and contours within a modified mask image according to an embodiment of the present invention.

[0123] Referring to FIG. 9, the computing device (300) can select three specific points (911, 912, 913) within the contour of the MRI image (91) and three specific points (921, 922, 923) within the contour (921) of the modified mask image (92).

[0124] In this case, the specific point may be a point arbitrarily designated by the user, or it may be a pre-set point.

[0125] For example, a specific point can be multiple points, such as both endpoints and specific parts.

[0126] And, the computing device (300) can calculate relative positions and distances, etc. based on the position coordinates of three specific points in the MRI image (91).

[0127] Additionally, the computing device (300) can calculate relative positions and distances based on the location coordinates of specific points within the modified mask image (92).

[0128] In addition, the computing device (300) can synthesize the images so that the location and distance of specific points of each calculated image correspond. At this time, the computing device (300) may also perform size adjustment and rotation of the modified mask image so that three specific points correspond to each other.

[0129] That is, the computing device (300) can generate a corrected MRI image (93) by synthesizing a first specific point (911) of the contour of the MRI image and a first specific point (921) of the contour in the modified mask image, a second specific point (912) of the contour in the MRI image and a second specific point (922) of the contour in the modified mask image, and a third specific point (913) of the contour in the MRI image and a second specific point (923) of the contour in the modified mask image.

[0131] And, the computing device (300) can provide the corrected MRI image to the user (S400).

[0132] Specifically, contour-corrected MRI images allow for the accurate identification of structures compared to conventional MRI images with disconnected contours.

[0133] In addition, when image segmentation is performed after post-processing MRI images, the Dice Score, a similarity metric for disconnected MRI images, is 0.55, whereas the Dice Score for contour-corrected MRI images can be higher at 0.824.

[0134] Accordingly, the computing device (300) can provide the user with an MRI image with corrected contours so that it can be used for diagnosis.

[0135] For example, contour-corrected MRI images can be utilized for diagnosis, such as surgical planning, radiation therapy, chemotherapy, or other customized treatment plans, as well as patient monitoring.

[0137] With reference to FIG. 10 below, a specific hardware implementation of a server as a computing device for performing the contour correction method according to the present embodiment will be described.

[0138] FIG. 10 is an exemplary diagram showing the configuration of a computing device that performs a contour correction method according to one embodiment of the present invention.

[0139] Referring to FIG. 10, in some embodiments of the present invention, the computing device (300) may be implemented in the form of a computing device. One or more of the modules constituting the computing device (300) are implemented on a general-purpose computing processor and thus may include a processor (388), an input / output I / O (382), a memory (384), an interface (386), and a bus (385). The processor (388), the input / output device (382), the memory (384), and / or the interface (386) may be coupled to each other through the bus (385). The bus (385) corresponds to a path through which data travels.

[0140] Specifically, the processor (388) may include at least one of a CPU (Central Processing Unit), MPU (Micro Processor Unit), MCU (Micro Controller Unit), GPU (Graphic Processing Unit), microprocessor, digital signal processor, microcontroller, application processor (AP), and logic elements capable of performing similar functions.

[0141] The input / output device (382) may include at least one of a keypad, a keyboard, a touchscreen, and a display device. The memory device (384) may store data and / or programs, etc.

[0142] The interface (386) can perform the function of transmitting data to a communication network or receiving data from a communication network. The interface (386) may be wired or wireless. For example, the interface (386) may include an antenna or a wired / wireless transceiver, etc. The memory (384) is a volatile operational memory for enhancing the operation of the processor (388) and protecting personal information, and may further include high-speed DRAM and / or SRAM, etc.

[0143] Additionally, the memory (384) stores programming and data configurations that provide the functions of some or all of the modules described herein. For example, it may include logic that enables the execution of selected embodiments of the contour correction method described above.

[0144] A computer program may be executed by a processor, comprising a set of instructions including each step of performing the above-described contour correction method stored in memory (384), an operation to generate a mask image for the contour of a structure within a program or MRI image, an operation to generate a skeleton for the contour within the generated mask image, an operation to modify the mask image through image processing using the skeleton, and an operation to correct the contour of a structure within an MRI image through image processing using the modified mask image.

[0145] According to the present invention, the invention has the effect of identifying structural features by correcting disconnected contours within an MRI image.

[0146] In addition, the present invention has the effect of enabling more accurate image segmentation after post-processing of MRI images by correcting disconnected contours within MRI images.

[0148] Furthermore, the various embodiments described herein may be implemented, for example, in a recording medium readable by a computer or similar device using software, hardware, or a combination thereof.

[0149] According to hardware implementation, the embodiments described herein may be implemented using at least one of ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), processors, controllers, microcontrollers, microprocessors, and other electrical units for performing functions. In some cases, the embodiments described herein may be implemented as the control module itself.

[0150] According to software implementation, embodiments such as the procedures and functions described herein may be implemented in separate software modules. Each of the software modules may perform one or more functions and operations described herein. Software code may be implemented as a software application written in a suitable programming language. The software code may be stored in a memory module and executed by a control module.

[0151] The above description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains will be able to make various modifications, changes, and substitutions within the scope of the essential characteristics of the present invention without departing from its nature.

[0152] Accordingly, the embodiments disclosed in this invention and the accompanying drawings are intended to illustrate, not limit, the technical concept of the invention, and the scope of the technical concept of the invention is not limited by such embodiments and accompanying drawings. The scope of protection of this invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of this invention.

Claims

Claim 1 A contour correction method for a structure within an MRI image performed on a computing device, comprising: a step of generating a mask image for the contour of a structure within an MRI image; a step of generating a skeleton for the contour within the generated mask image; a step of modifying the mask image through image processing using the skeleton; and a step of correcting the contour of the structure within the MRI image through image processing using the modified mask image. Claim 2 A contour correction method according to claim 1, wherein the step of generating the skeleton is characterized by expanding the contours within the mask image and connecting the inner centers of the expanded contours to generate the skeleton. Claim 3 A contour correction method according to claim 1, wherein the modifying step is characterized by modifying disconnected contours within the mask image through image processing that composites the skeleton with the mask image. Claim 4 A contour correction method according to claim 1, wherein the correction step corrects disconnected contours within the MRI image through image processing that synthesizes a modified mask image into the MRI image. Claim 5 A contour correction method characterized by further including the step of generating a contour for a structure within an MRI image using an artificial neural network model, in claim 1. Claim 6 A contour correction method according to claim 5, further comprising a step of determining whether the generated contour is disconnected; wherein the step of generating a mask image is characterized by generating a mask image for the disconnected contour. Claim 7 A contour correction method according to claim 6, wherein the determining step is determined based on the number of regions of interest adjacent to a specific part based on a specific part of the contour. Claim 8 A computing device comprising: a processor; a memory for loading a computer program executed by the processor; and a storage for storing the computer program, wherein the computer program comprises: an operation to generate a mask image for the contour of a structure in an MRI image; an operation to generate a skeleton for the contour in the generated mask image; an operation to modify the mask image through image processing using the skeleton; and an operation to correct the contour of the structure in the MRI image through image processing using the modified mask image. Claim 9 A computing device according to claim 8, wherein the operation of generating the skeleton is characterized by expanding the contours within the mask image and connecting the inner centers of the expanded contours to generate the skeleton. Claim 10 A computing device according to claim 8, wherein the modifying operation modifies disconnected contours within the mask image through image processing that composites the skeleton with the mask image. Claim 11 A computing device according to claim 8, wherein the correcting operation corrects disconnected contours within the MRI image through image processing that synthesizes a modified mask image into the MRI image. Claim 12 A computing device characterized by further including, in claim 8, the operation of generating contours for structures within an MRI image using an artificial neural network model. Claim 13 A computing device according to claim 12, further comprising an operation to determine whether the generated contour is disconnected; wherein the operation to generate a mask image is characterized by generating a mask image for the disconnected contour. Claim 14 A computing device according to claim 13, wherein the determining operation is determined based on the number of regions of interest adjacent to a specific part based on a specific part of the contour. Claim 15 A computer-readable recording medium storing a program that performs a contour correction method described in any one of claims 1 to 7. Claim 16 A computer program containing program code for executing a contour correction method described in any one of claims 1 to 7, stored on a computer-readable recording medium.

Images