Method and device for medical image conversion using artificial intelligence in the frequency domain

Abstract

The present invention relates to a medical image conversion method, and more particularly, to an image conversion method for converting medical images using a generative adversarial network (GAN). According to one embodiment of the present invention, there is provided a medical image conversion method that enables medical staff to perform medical examinations, diagnoses, and treatments with accurate information through conversion between medical images using a learning model that performs machine learning, thereby enabling patients to receive appropriate medical services.

Description

[Technical Field]

[0001] One aspect of the present invention relates to a method and apparatus for converting medical images, and more particularly, to a method and apparatus for converting medical images using a generative adversarial network (GAN) in the frequency domain. [Background technology]

[0002] The material described in this section merely provides background information for embodiments of the present invention and may not constitute prior art.

[0003] Non-enhanced CT images, which do not use contrast enhancers, and enhanced CT images, which use contrast enhancers, are widely used for the diagnosis and treatment of patients with emergency cerebral hemorrhage and tumors.

[0004] For example, contrast agents that increase tissue contrast are useful for detecting blood vessels, various organs, cancer, etc. However, contrast agents often cause side effects, and in severe cases, can lead to death due to cardiac arrest, shock, etc.

[0005] Since such side effects cannot be predicted in advance, image conversion can be used to convert non-contrast-enhanced CT or MR images into contrast-enhanced CT or MR images, thereby preventing such side effects in advance and reducing medical costs associated with contrast agents.

[0006] In addition, in cases where MR images are available but X-ray irradiation is not possible and CT images are not available, such as pregnant women, or in cases where CT images are available but MR images cannot be taken due to time, cost, or implants, image conversion between imaging modalities still allows medical staff to determine tissue boundaries or the location of tumors without the use of contrast agents.

[0007] In addition, it can help medical staff better understand information about lesions through conversion between MR images, for example, conversion from T1 image to T2 image, or conversion from T1 image to diffusion image.

[0008] Meanwhile, a typical learning method for a learning model is that it receives an input image, synthesizes an output image based on it, and then compares the synthesized output image with a target image, and learning proceeds in the direction of reducing the difference.

[0009] In this way, the learning of a general learning model generates an output image in the image domain, but if the frequency form between the image in the frequency domain of this output image and the image in the frequency domain of the target image matches, for example, if the high frequency components match, the output image can be output more vividly. Therefore, it is worthwhile to research an artificial intelligence learning model in the frequency domain.

[0010] The above-mentioned background art is technical information that the inventor possessed for the purpose of deriving the embodiments of the present invention or that he acquired in the process of deriving the embodiments of the present invention, and cannot necessarily be said to be publicly known art that was disclosed to the general public prior to the filing of the embodiments of the present invention. Summary of the Invention [Problem to be solved by the invention]

[0011] In order to solve the above-mentioned problems, one aspect of the present invention is to provide a medical image conversion method and apparatus using a learning model that converts all of the output image and target image into images in the frequency domain in order to obtain a clear output image, and then learns to reduce the difference by comparing them in the frequency domain.

[0012] Another object of the present invention is to provide a medical image conversion method that enables medical staff to perform medical examinations, diagnoses, and treatments with accurate information through conversion between medical images, thereby enabling patients to receive appropriate medical services.

[0013] The technical problems that the present invention aims to achieve are not limited to the technical problems mentioned above, and other technical problems not mentioned above will be clearly understood by those skilled in the art to which the present invention pertains from the following description. [Means for solving the problem]

[0014] In order to achieve the above-mentioned object, one aspect of the present invention is a medical image conversion method for converting medical images using a generative adversarial network (GAN), comprising:

[0015] a first step of receiving an input of a first image selected from a paired dataset including a first image and a second image;

[0016] The medical image conversion method may include one or more of: a second step of generating a third image corresponding to the first image based on the first image; and a third step of converting the second image and the third image into images in the frequency domain, comparing the images, and reflecting the results in a learning model that performs medical image conversion.

[0017] According to an embodiment, in the third step, the image on the frequency domain may include an image on K-Space.

[0018] In some embodiments, the third step includes comparing a second frequency image obtained by frequency-transforming the second image with a third frequency image obtained by frequency-transforming the third image, and reflecting the resultant value in the learning model (a learning model that learns in the frequency domain) to derive an improved third frequency image; and

[0019] and inversely transforming the improved third frequency image into an image in the image domain.

[0020] Another aspect of the present invention is a medical image conversion device that converts medical images using a generative adversarial network (GAN),

[0021] A first learning model that receives an input of a first image selected from a paired dataset including a first image and a second image and generates a third image based on the first image; and

[0022] A medical image conversion device can be provided that includes a second learning model that receives an input of a third frequency image in which the third image is converted into an image in the frequency domain, and generates a third frequency image with improved image quality based on the third frequency image.

[0023] According to an embodiment, the second image may be converted into an image in the frequency domain, and the third image may be frequency-converted to obtain a second frequency image. The second image may be compared with the third frequency image, and the difference between the second image and the third image may be fed back to the second learning model.

[0024] According to an embodiment, the image processing apparatus may further include an inverse transform unit for inversely transforming the improved third frequency image to generate an improved third image.

[0025] According to an embodiment, the third frequency image, the second frequency image, and the image quality-improved third frequency image may be images on K-Space.

[0026] Another aspect of the present invention is a medical image conversion method for converting medical images using a generative adversarial network (GAN),

[0027] an image generating step of generating a third image in response to an input of a first image;

[0028] a first frequency transformation step of transforming the third image into a third frequency image in the frequency domain; and

[0029] and generating an improved third frequency image in response to the third frequency image.

[0030] a second frequency transformation step of transforming a second image constituting a paired data set with the first image into a second frequency image in the frequency domain according to an embodiment; and

[0031] The method may further include an image comparing step of comparing the second frequency image with the third frequency image and reflecting the difference between the second frequency image and the third frequency image in the image quality improving step.

[0032] According to an embodiment, the method may further include an inverse transform step of inverse frequency transforming the improved third frequency image to output the improved third image. [Effects of the Invention]

[0033] As described above, according to one embodiment of the present invention, a method and apparatus for medical image conversion using a learning model that converts all output images and target images into images in the frequency domain in order to obtain a clear output image and then learns to reduce the difference by comparing them in the frequency domain is provided.

[0034] In addition, according to another embodiment of the present invention, there is provided a medical image conversion method that enables medical staff to perform medical examinations, diagnoses, and treatments with accurate information through conversion between medical images, thereby enabling patients to receive appropriate medical services.

[0035] In addition, the present invention has various effects such as excellent versatility depending on the embodiment, and such effects can be clearly confirmed in the description of the embodiments below. [Brief explanation of the drawings]

[0036] The following drawings attached to this specification illustrate one embodiment of the present invention and, together with the detailed description of the invention described above, serve to further understand the technical concept of the present invention, and therefore the present invention is not limited to the details shown in such drawings.

[0037] [Figure 1] FIG. 1 illustrates a medical image conversion method according to one embodiment of the present invention. [Figure 2] FIG. 2 illustrates an artificial intelligence learning model according to one embodiment of the present invention. [Figure 3] FIG. 3 shows an embodiment of calculating the match rate between the output image and the target image. [Figure 4] FIG. 4 illustrates a medical image conversion method according to another embodiment of the present invention. [Figure 5] FIG. 5 illustrates a medical image conversion method according to another embodiment of the present invention. [Figure 6] FIG. 6 illustrates an artificial intelligence learning model according to another embodiment of the present invention. [Figure 7] FIG. 7 illustrates a medical image conversion method according to another embodiment of the present invention. [Figure 8] FIG. 8 illustrates a medical image conversion method according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0038] The advantages and features of the present invention, and methods for achieving them, will become clearer with reference to the following detailed description of the embodiments accompanied by the accompanying drawings. However, the present invention is not limited to the following embodiments, but may be embodied in various different forms, and should be understood to include all modifications, equivalents, and alternatives within the spirit and technical scope of the present invention. The following embodiments are provided to ensure a complete disclosure of the present invention and to fully convey the scope of the invention to those skilled in the art. In describing the present invention, if a detailed description of related prior art is considered to obscure the gist of the present invention, such a detailed description will be omitted.

[0039] The terms used in this application are merely used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise.

[0040] In this application, the terms "comprise" or "have" are intended to specify the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not preclude the presence or additional possibility of one or more other features, numbers, steps, operations, components, parts, or combinations thereof. Terms such as "first," "second," etc. may be used to describe various components, but the components should not be limited by these terms. These terms are used only to distinguish one component from another.

[0041] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description with reference to the accompanying drawings, identical or corresponding components will be given the same drawing numbers, and duplicate descriptions thereof will be omitted.

[0042] FIG. 1 illustrates a medical image conversion method according to an embodiment of the present invention, and FIG. 2 illustrates an artificial intelligence learning model according to an embodiment of the present invention.

[0043] According to one embodiment of the present invention, there is provided a medical image conversion method for converting medical images using a generative adversarial network (GAN).

[0044] The medical image conversion method according to the present embodiment includes an input step (S100) of receiving a first image 110 selected from a paired data set including a first image 110 and a second image 120; a generation step (S110) of generating a third image 130 corresponding to the first image 110 and having a domain different from that of the first image 110 based on the first image 110;

[0045] The method may further include a learning step (S120) of comparing the third image 130 with the second image 120 and learning a learning model in consideration of the comparison result.

[0046] The first image 110 and the second image 120 may represent medical images. Here, the medical images may include magnetic resonance (MR) images, such as 2D MR, 3D MR, 2D streaming MR, 4D MR, 4D volumetric MR, and 4D cine MR; functional MR images, such as fMR, DCE-MR, and diffusion MR; computed tomography (CT) images, such as 2D CT, conebeam CT, 3D CT, and 4D CT; ultrasound images, such as 2D ultrasound, 3D ultrasound, and 4D ultrasound; positron emission tomography (PET) images; X-ray images; fluoroscopic images; radiotherapy portal images; single-photon emission computed tomography (SPECT) images; and computer-generated synthetic images, such as pseudo-CT.

[0047] In addition, the medical image may include medical image data, such as a training image, a ground truth image, a contoured image, a dose image, and the like.

[0048] Depending on the embodiment, the medical image may include an image acquired from an image acquisition device, a computer-generated image, etc. The image acquisition device may include, for example, an MR imaging device, a CT imaging device, a PET imaging device, an ultrasound imaging device, a fluoroscopic device, a SPECT imaging device, an integrated linear accelerator and MR imaging device, etc. In addition, the image acquisition device is not limited to the above examples and may include various medical imaging devices within the scope of the technical concept for acquiring a medical image of a patient.

[0049] The paired data set may include a set of training data including images of an object, for example, a specific part of a patient.

[0050] A CT image and an MR image can be taken of a specific part of a patient, for example, the head, using a CT imaging device and an MR imaging device, respectively, and in this case, the CT image and the MR image can form a paired data set.

[0051] When non-enhanced CT images and MR images are taken of a specific area of ​​a patient, the non-enhanced CT images and MR images can form a paired data set, and when contrast-enhanced CT images and MR images are taken of a specific area of ​​a patient, the contrast-enhanced CT images and MR images can form a paired data set.

[0052] When T1 and T2 images are taken of a specific part of a patient, the T1 and T2 images can form a paired data set.

[0053] Here, the specific part of a patient may refer to a specific part of the same patient, or may refer to a specific part of different patients if the parts being imaged are the same.

[0054] For example, CT images and MR images of the same specific area of ​​a patient can form a paired data set, and CT images and MR images of the same area of ​​different patients can also form a paired data set.

[0055] What does it mean for two images to have different domains? For example, if the pattern contours contained in the two corresponding images differ between the two images, or if the intensity distribution or pattern between the two images is different, then the domains are different.

[0056] In addition, images taken with different imaging devices have different domains. For example, CT images and MR images are taken with CT imaging devices and MR imaging devices, respectively, so they have different domains. Similarly, there are different domains between ultrasound images and CT images, between ultrasound images and MR images, and between MR images and PET images.

[0057] Furthermore, non-contrast-enhanced images and contrast-enhanced images have different intensity distributions for the same region, and therefore have different domains.

[0058] In the case of MR images, T1 images and T2 images are the same MR images, but they have different intensities and therefore different domains. In other words, if there is a difference in the signal, sequence, or image creation process that is the basis for image creation between two images created through reconstruction based on signals provided by an individual, the two images with the difference will have different domains. In addition, the domains of the examples of medical images mentioned above are different.

[0059] To explain a medical image conversion method according to an embodiment of the present invention, an example will be described in which an image acquisition device acquires a first image 110 and a second image 120 of the same part of the same patient and obtains paired data sets. In this embodiment, the first image 110 is an original CT image, the second image 120 is an original MR image, and the third image 130 is a combined MR image.

[0060] The input step (S100) may include a step of receiving, as input to a learning model, a first image 110 selected from a paired dataset including a first image 110 and a second image 120. Here, the first image 110 may include an input image obtained by processing an original CT image using learning data.

[0061] The manufacturer of the software medical device that performs the medical image conversion method according to this embodiment can determine which of the first image 110 and the second image 120 to select as the input image in the paired data set including the first image 110 and the second image 120. In this case, the image to be selected can be automatically selected by a selection algorithm or manually selected by an operator.

[0062] A manufacturer can create a paired dataset by acquiring original CT images using a CT imaging device and original MR images using an MR imaging device, or can select the original CT images as input images in the paired dataset using an algorithm or manually.

[0063] The generation step (S110) may include a step in which a learning model that receives an input of a first image 110 generates a third image 130 based on the first image 110, corresponding to the first image 110 and having a different domain from the first image 110.

[0064] The learning model can receive the original CT image as input and generate a synthetic MR image based on the original CT image.

[0065] The process in which the learning model generates the third image 130 from the first image 110, which has a domain different from that of the first image 110, may be performed through the following two processes according to an embodiment. However, the process in which the learning model generates the third image 130 is not limited to the following two processes.

[0066] The first process is composed of one or more layers that receive a first image 110 acquired by an image acquisition device and perform convolution operations to extract high-dimensional features smaller than the first image 110, and this can be expressed as an artificial neural network that extracts features. Such an artificial neural network that extracts features receives input of 2D or 3D actual image data acquired by an image acquisition device and performs convolution operations to extract image features from the original image.

[0067] To exemplarily explain this process, a convolution layer extracts image features through a filter and a pooling layer strengthens the features and reduces the size of the image. Image features can be extracted by repeating convolution and pooling.

[0068] In a convolution layer configured with multiple layers according to an embodiment, the first convolution layer may extract features of an input image provided as an input and output a feature map as a result, the pooling layer may receive the feature map as an input and output a result in which the features are enhanced and the image size is reduced, and the output of the pooling layer may again be input to the second convolution layer.

[0069] The second step is composed of one or more layers that perform deconvolution operations to generate image data of different modalities from the result of the first step, and can be expressed as an artificial neural network that generates images of different modalities. Such an artificial neural network that generates images of different modalities can generate two-dimensional or three-dimensional image data of different modalities by performing deconvolution operations from the results of the convolution operation performed by the artificial neural network that extracts features in the first step.

[0070] The learning step (S120) may include a step of comparing the third image 130 with the second image 120 and training a learning model based on the comparison results. If the original CT image is determined as the input image from among the original CT image and the original MR image included in the paired dataset, the original MR image may be determined as the target image. Here, the second image 120 may correspond to ground truth. After generating the third image 130, the learning model may compare the generated third image 130 with the second image 120 and perform learning to reduce the difference between the two. That is, the learning model may continuously compare the synthetic MR image with the original MR image to determine whether the synthetic MR image has been generated to be close to the original MR image, thereby training the synthetic MR image to be the same as the original MR image. According to an embodiment, the method may include a step of calculating a loss value between the third image 130 and the second image 120 by applying a loss function to the algorithm, and updating the parameters of the learning model based on the loss value.

[0071] According to an embodiment, in the step of calculating the loss value, the calculation of the loss value for updating the learning model may be performed for each iteration. That is, the learning model may calculate the loss value between the synthetic MR image and the original MR image for each iteration while performing learning.

[0072] When the number of repetitions includes a first repetition interval and a second repetition interval that are connected sequentially, depending on the embodiment, if the absolute value of the change between the first loss value calculated in the first repetition interval and the second loss value calculated in the second repetition interval is smaller than a reference value, the learning model may regard this as overfitting and terminate learning.

[0073] Once the learning model has sufficiently learned medical image conversion, it can complete the learning without further learning. The learning model that has completed the learning will now have the ability to convert medical images into third images 130 that have a different domain from the first images 110 for various first images 110 each having a different data distribution.

[0074] According to an embodiment of the present invention, the learning model 200 may be a GAN model and may include a generator and a classifier. The generator may perform image generation learning, and the classifier may perform image segmentation learning.

[0075] Referring to FIG. 2, in a learning model 200 to which a GAN model is applied, when an original CT image 110 is input as an input image to a generator 210, the generator 210 performs learning to generate a synthetic MR image 130 through the above-mentioned convolution operation, and a classifier 220 can perform learning to distinguish between the original MR image 120, i.e., the actual medical image, and the synthetic MR image 130.

[0076] According to an embodiment, a cycle-GAN model (not shown) may be applied to the learning model 200 according to this embodiment. In this case, the learning model 200 may be configured to include a first generator 210, a first division molecule 220, a second generator, and a second division molecule. When the original CT image 110 is input as an input image to the first generator 210, the first generator 210 performs learning to generate a synthetic MR image 130 through the convolution operation described above, and the first division molecule 220 may perform learning to distinguish between the original MR image 120, i.e., the actual medical image, and the synthetic MR image. Furthermore, when the synthetic MR image 130 generated by the first generator 210 is input to the second generator, the second generator may perform learning to generate a synthetic CT image based on the synthetic MR image 130, and the second division molecule may perform learning to distinguish the synthetic CT image from the original CT image.

[0077] Here, the original CT image 110 corresponds to the first image 110 in the above-mentioned medical image conversion method, the synthesized MR image 130 corresponds to the third image 130, and the original MR image 120 corresponds to the second image 120.

[0078] According to an embodiment, the learning step (S120) may include a step of calculating a match rate between the third image 130 and the second image 120 and feeding the calculated value back to the learning model. Here, calculating the match rate is a method for determining the degree of difference between two images belonging to the same domain.

[0079] FIG. 3 shows an embodiment of calculating the match rate between the output image and the target image.

[0080] In some embodiments, if a lesion area is displayed in the same location on the composite MR image and the original MR image, the matching rate can be calculated by comparing the size of the corresponding lesion area. For example, the positions of tumors and other lesions can be displayed on the composite MR image and the original MR image, respectively, and then converted to binary images to calculate the matching rate.

[0081] In this case, the match rate can be calculated using evaluation indices such as the Hausdorff distance or the Dice similarity coefficient, or can be quantitatively calculated based on the difference in the centers of mass of the two lesions. If the target and output images are CT images, the match rate between the two images can be determined by comparing the calculated radiation doses for the lesions. Depending on the embodiment, the difference between the synthesized and original MR images can be quantified using MAE, RMSE, SSIM, PSNR, etc. (hereinafter referred to as performance indices) for determination.

[0082] Referring to Figure 3, the process of aligning the lesion area between the synthesized MR image 130, which is the output image, and the original MR image 120, which is the target image, and then applying the Hausdorff distance measurement method to compare the contours of each lesion and calculate the matching rate is shown.

[0083] FIG. 4 shows a medical image conversion method according to another embodiment of the present invention, and FIG. 5 shows a medical image conversion method according to yet another embodiment of the present invention.

[0084] FIG. 6 shows an artificial intelligence learning model according to another embodiment of the present invention.

[0085] A medical image conversion method according to an embodiment of the present invention is applied to convert medical images using a generative adversarial network (GAN).

[0086] The medical image conversion method according to this embodiment includes a first step (S200) of receiving an input of the first image 110 selected from a paired data set including the first image 110 and the second image;

[0087] A second step (S210) of generating a third image corresponding to the first image 110 based on the first image 110; and

[0088] and a third step (S220) of converting the second image and the third image into images in the frequency domain, comparing the results, and reflecting the results in a learning model 300 that performs medical image conversion. The learning model 300 referred to here may refer to a learning model that learns in the frequency domain. This will be referred to as the second learning model 300 in the following embodiments.

[0089] In this embodiment, the first image 110 may represent an input image, which is learning data input to a learning model, the second image 120 may represent a target image, and the third image 130 may represent an output image.

[0090] According to an embodiment, in the third step (S220), the image in the frequency domain may include an image in k-space.

[0091] In this specification, k-Space may correspond to the image domain.

[0092] For example, during MR imaging, if you apply an RF pulse and then change the magnitude of the phase-encoding gradient (G) step by step, you can obtain signals containing a lot of positional information. This data is called raw data. This data contains both positional and contrast information, and K-space can refer to a collection of raw data that can be used to create an image.

[0093] K-space stores all image information in the frequency domain, meaning that distorting or losing certain parts of K-space will inevitably result in changes in the image.

[0094] For example, when comparing the K-space where high frequency components have been lost with the image domain that matches it, we can see that the image in the image domain becomes slightly blurred as the high frequency components corresponding to the corners in the K-space are increasingly lost. This is a phenomenon that occurs when the boundary areas in the image that correspond to the high frequency components are broken.

[0095] In addition, when comparing the K-space with its matching image domain, where low-frequency components have been lost, it can be seen that the low-frequency components corresponding to the center of the K-space are increasingly lost, leaving only the boundary areas, which are the high-frequency components of the image, and light and dark and contrast are being destroyed.

[0096] In this invention, K-space can contain all information in the image domain. In other words, the designer can obtain the desired image shape depending on how the information in K-space is used. If you want an image with only edge components, you can simply remove low-frequency components, and if you want a smoothing effect, you can simply remove high-frequency components.

[0097] According to an embodiment, the third step (S220) may include a step (S221) of comparing the frequency-converted second frequency image 121 of the second image with the frequency-converted third frequency image 131 of the third image and reflecting the resulting value in a learning model to derive an improved third frequency image 301, and a step (S222) of inversely converting the 'improved third frequency image' 301 into an image in the image domain.

[0098] Finally, the 'improved third frequency image' 301 can be inversely transformed into an image in the image domain to output the 'improved third image' 132. The 'improved third image' is an image in the image domain like the third image 130, but since it is an output image output by the learning model 300 after performing learning in the frequency domain, it can show an improved result compared to the output image output by the learning model 200 that simply learns in the image domain. Here, improved can, for example, mean that the image clarity has increased.

[0099] The third frequency image 131, which is an image in the frequency domain, is compared with the second frequency image, and the calculated value is fed back to the learning model 300, which learns in the frequency domain, so that the learning model 300 matches the third frequency image with the second frequency image, thereby making it possible to match the third image 130 with the second image 120 in the image domain. In other words, by matching the frequency shapes in the frequency domain, it is possible to more precisely match the images in the image domain.

[0100] Another aspect of the present invention is a medical image conversion device that converts medical images using a generative adversarial network (GAN), comprising:

[0101] A first learning model 200 that receives an input of the first image 110 selected from a paired dataset including the first image 110 and a second image and generates a third image based on the first image 110; and

[0102] and a second learning model 300 that receives a third frequency image 131 obtained by converting a third image into an image in the frequency domain and generates a third frequency image 301 with improved image quality based on the third frequency image 131. Here, the first learning model 200 may include a model that learns on the image domain, and the second learning model 300 may include a model that learns on the frequency domain.

[0103] The medical image conversion device according to this embodiment may further include a comparison unit 310 that compares a second frequency image 121 obtained by converting the second image into an image in the frequency domain with a third frequency image 131 obtained by frequency converting the third image, and feeds back the difference to the second learning model 300.

[0104] According to this embodiment, the comparison unit 310 performs a function of comparing the frequency forms of the second frequency image 121 and the third frequency image 131 in the frequency domain.

[0105] According to an embodiment, the medical image conversion device may further include an inverse conversion unit that inversely converts the image quality improved third frequency image 301 to generate the image quality improved third image 132. Here, the image quality improved third image 132 may include an image with improved image clarity.

[0106] According to an embodiment, the third frequency image 131, the second frequency image 121, and the image quality-enhanced third frequency image 301 may be images in K-space.

[0107] 7 and 8 show a medical image conversion method according to another embodiment of the present invention.

[0108] Meanwhile, another aspect of the present invention is a medical image conversion method for converting medical images using a generative adversarial network (GAN),

[0109] an image generating step (S300) of receiving the first image 110 and generating a third image 130;

[0110] a first frequency conversion step (S310) of converting the third image into a third frequency image 131 in the frequency domain; and

[0111] and an image enhancement step (S320) of receiving the third frequency image 131 and generating an improved third frequency image 301.

[0112] Here, the image quality improvement step (S320) may be performed by a learning model 300 that learns in the frequency domain. That is, the learning model 300 may receive the third frequency image 131 as an input and generate an improved third frequency image 301 as an output.

[0113] A second frequency transformation step (S330) of transforming a second image constituting a paired data set with the first image 110 into a second frequency image 121 in the frequency domain according to an embodiment; and

[0114] The method may include an image comparison step (S340) of comparing the second frequency image 121 with the third frequency image 131 and reflecting the difference in the image quality enhancement step.

[0115] In some embodiments, the method may include an inverse transform step (S350) of inversely transforming the improved third frequency image 301 to output an 'improved third image' 132. Here, the 'improved third image' 132 may refer to an image with improved image quality as an image in the image domain.

[0116] The above-described embodiments of the present invention may be implemented in the form of a computer program that can be executed by various components on a computer, and such a computer program may be recorded on a computer-readable medium, which may include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program instructions, such as ROMs, RAMs, and flash memories.

[0117] Meanwhile, the computer program may be one specially designed and constructed for the present invention, or one that is known and available to those skilled in the art of computer software. Examples of the computer program include not only machine language code, such as that produced by a compiler, but also high-level language code that can be executed by a computer using an interpreter, etc.

[0118] In the present specification (particularly in the claims), the use of the term "said" and similar indicators may be applicable to both the singular and the plural. Furthermore, when a range is described in the present invention, it is considered to include inventions to which individual values ​​within the range are applied (unless otherwise specified), and is the same as if each individual value constituting the range were described in the detailed description of the invention.

[0119] Unless explicitly stated or stated to the contrary, steps constituting the method of the present invention may be performed in any suitable order. The present invention is not necessarily limited to the order of steps described above. The use of all examples or exemplary terms (e.g., etc.) in the present invention is merely for the purpose of describing the present invention in detail, and the scope of the present invention is not limited by the claims. Furthermore, those skilled in the art will recognize that various modifications, combinations, and variations can be made within the scope of the appended claims or their equivalents, depending on design conditions and factors.

[0120] Therefore, the concept of the present invention should not be limited to the above-described embodiments, and not only the scope of the claims described below, but also all scopes equivalent to or modified equivalently from the scope of the claims belong to the scope of the concept of the present invention. [Explanation of symbols]

[0121] 100 AI learning models 110 First Image 120 Second Image 121 Second Frequency Image 130 Third Image 131 Third Frequency Image 132 Improved Third Image 200 First Learning Model 300 Second Learning Model 301 Improved 3rd Frequency Image 310 Comparison Section

Claims

1. A medical image conversion method for converting medical images using a generative adversarial network (GAN), a first step of receiving an input of a first image selected from a paired dataset including a first image and a second image; a second step of generating a third image corresponding to the first image based on the first image; and and a third step of converting the second image and the third image into images in the frequency domain, comparing the images, and reflecting the results of the comparison in a learning model for performing medical image conversion. A medical image conversion method comprising:

2. In the third step, the image in the frequency domain includes an image in K-Space. The medical image conversion method according to claim 1 .

3. The third step includes comparing a second frequency image obtained by frequency-transforming the second image with a third frequency image obtained by frequency-transforming the third image, and reflecting the resultant value in the learning model (a learning model that learns in the frequency domain) to derive an improved third frequency image; and and transforming the improved third frequency image back into an image in the image domain. The medical image conversion method according to claim 1 .

4. A medical image conversion device that converts medical images using a generative adversarial network (GAN), A first learning model that receives an input of a first image selected from a paired dataset including a first image and a second image and generates a third image based on the first image; and a second learning model that receives an input of a third frequency image obtained by converting the third image into an image in the frequency domain and generates a third frequency image with improved image quality based on the third frequency image. A medical image conversion device characterized by:

5. a comparison unit that compares a second frequency image obtained by converting the second image into an image in the frequency domain with a third frequency image obtained by frequency-converting the third image, and feeds back a difference between the second frequency image and the third frequency image to the second learning model.

5. The medical image conversion device according to claim 4.

6. an inverse transform unit for inversely transforming the improved third frequency image to generate an improved third image; 6. The medical image conversion device according to claim 5.

7. The third frequency image, the second frequency image, and the image quality-improved third frequency image are images in K-Space.

5. The medical image conversion device according to claim 4.

8. A medical image conversion method for converting medical images using a generative adversarial network (GAN), an image generating step of receiving an input of a first image and generating a third image; a first frequency transform step of transforming the third image into a third frequency image in the frequency domain; and and generating an improved third frequency image by receiving the third frequency image. A medical image conversion method comprising:

9. a second frequency transformation step of transforming a second image that constitutes a paired data set with the first image into a second frequency image in the frequency domain; and and an image comparing step of comparing the second frequency image with the third frequency image and reflecting the difference in the image quality improving step. The medical image conversion method according to claim 8.

10. and an inverse transform step of inverse frequency transforming the improved third frequency image to output the improved third image. The medical image conversion method according to claim 9.

Images