Medical image restoration and transmission system and method

Abstract

The present invention provides a medical image restoration and transmission method using a medical image restoration and transmission system, the method including: receiving medical image data including at least one of accelerated k-space data and DICOM data generated based on the accelerated k-space data; restoring the received medical image data using an artificial neural network model; and transmitting the restored medical image data based on an address where the medical image data was received.

Description

Technical Field

[0001] The present invention relates to a medical video restoration and transmission system and method that restores medical videos using artificial intelligence, performs encryption and decryption, and transmits medical videos.

Background Art

[0002] Generally, in a hospital, medical videos taken by medical video shooting and diagnostic devices such as X-ray, CT, MRI, or ultrasonic waves are acquired and stored as digital data, and the functions necessary for transmitting, searching, etc. the interpretation and medical records to each terminal are integrally processed using a digital medical video storage and communication system (PACS: picture archiving communication system). PACS stores and manages medical images to conform to DICOM (Digital Imaging and Communications in Medicine), an industry-standard communication protocol for transmitting medical images and information between different types of medical imaging devices.

[0003] In addition, PACS enables the processing of hospital operations that were previously diagnosed and interpreted using existing films via computers and networks, thereby enabling efficient hospital operations.

[0004] On the other hand, as artificial intelligence develops, technologies have been developed to enable artificial intelligence such as deep learning algorithms to be learned using a large amount of medical video data and utilized for medical judgment.

[0005] However, the medical video data used in PACS cannot be immediately used for learning deep learning algorithms, nor can it be immediately applied for restoration and processing using deep learning algorithms. Therefore, it is necessary to develop a system capable of preprocessing or postprocessing.

[0006] In addition, in the transmission and reception of medical video data between a PACS server and an external server that provides deep learning algorithms, there are security-sensitive issues such as personal information and body images contained in the medical video data. Therefore, encryption such as anonymization of the medical video data is necessary.

Summary of the Invention

Problems to be Solved by the Invention

[0007] A medical video restoration and transmission method according to an embodiment of the present invention anonymizes and encrypts medical video data including personal information transmitted and received between a medical institution and a medical video restoration and transmission system, and provides enhanced security to prevent personal information from leaking.

[0008] A medical video restoration and transmission method according to an embodiment of the present invention aims to obtain medical videos of improved quality by restoring medical videos using an optimal artificial neural network model according to the characteristic information of the medical video data.

Means for Solving the Problems

[0009] An embodiment of the present invention is a medical video restoration and transmission method using a medical video restoration and transmission system, including receiving medical video data including at least one of accelerated k-space data and DICOM data generated based on the accelerated k-space data, restoring the received medical video data using an artificial neural network model, and transmitting the restored medical video data based on the address where the medical video data was received.

[0010] In this embodiment, a medical video restoration and transmission method is provided in which the medical video data is encrypted before being restored using the artificial neural network model.

[0011] In this embodiment, the encrypting includes shuffling a plurality of video slices included in the medical video data with each other, and aims to provide a medical video restoration and transmission method.

[0012] In this embodiment, it aims to provide a medical video restoration and transmission method that stores the encrypted medical video data and deletes the unencrypted medical video data.

[0013] In this embodiment, it aims to provide a medical video restoration and transmission method that decrypts the stored encrypted medical video data and restores the decrypted medical video data by means of the artificial neural network model.

[0014] In this embodiment, it aims to provide a medical video restoration and transmission method that classifies the medical video data according to the characteristic information of the medical video data and transmits it to the artificial neural network model, and distributes and transmits the medical video data classified by the characteristic information to a plurality of pre-classified artificial neural network models based on the characteristic information for input.

[0015] In this embodiment, the characteristic information aims to provide a medical video restoration and transmission method that includes at least one of protocol information and sequence information of the medical video data.

[0016] In this embodiment, when the characteristic information includes both the protocol information and the sequence information, a protocol video set is generated from the medical video data based on the protocol information, a plurality of medical videos included in the protocol video set are classified as the sequence information, a sequence video set is generated based on the classified sequence information, and the sequence set is transmitted to the artificial neural network model, aiming to provide a medical video restoration and transmission method.

[0017] One embodiment of the present invention is a medical video restoration and transmission system that receives medical video data including at least one of accelerated k-space data and DICOM data generated based on the accelerated k-space data, restores the received medical video data using an artificial neural network model, and transmits the restored medical video data based on the address where the medical video data was received.

[0018] In this embodiment, a medical video restoration and transmission system is provided that encrypts the medical video data before restoring it using the artificial neural network model.

[0019] In this embodiment, providing a medical video restoration and transmission system, the encrypting includes shuffling a plurality of video slices included in the medical video data with each other.

[0020] In this embodiment, a medical video restoration and transmission system is provided that stores the encrypted medical video data and deletes the medical video data that was not encrypted.

[0021] In this embodiment, a medical video restoration and transmission system is provided that decrypts the stored encrypted medical video data and restores the decrypted medical video data using the artificial neural network model.

[0022] In this embodiment, a medical video restoration and transmission system is provided that classifies the medical video data according to the characteristic information of the medical video data and transmits it to the artificial neural network model, and distributes and transmits the medical video data classified by the characteristic information to a plurality of pre-classified artificial neural network models based on the characteristic information for input.

[0023] In this embodiment, the medical video restoration and transmission system is to be provided, where the characteristic information includes at least one of protocol information and sequence information of medical video data.

[0024] In this embodiment, when the characteristic information includes both the protocol information and the sequence information, a protocol video set is generated from the medical video data based on the protocol information, a plurality of medical videos included in the protocol video set are classified as the sequence information, a sequence video set is generated based on the classified sequence information, and the sequence set is transmitted to the artificial neural network model, so as to provide a medical video restoration and transmission system.

Advantages of the Invention

[0025] The medical video restoration and transmission method according to an embodiment of the present invention pseudonymizes and encrypts medical video data including personal information transmitted between a medical institution and a medical video restoration and transmission system, so as to provide enhanced security, thereby preventing the leakage of personal information.

[0026] The medical video restoration and transmission method according to an embodiment of the present invention restores medical videos by using an optimal artificial neural network model according to the characteristic information of the medical video data, so as to obtain medical videos of improved quality.

Brief Description of the Drawings

[0027]

Figure 1

Figure 2

Figure 3

Figure 4a

Figure 4b

Figure 4c

Figure 5

Mode for Carrying Out the Invention

[0028] Hereinafter, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those having ordinary knowledge in the technical field to which the present invention pertains can easily implement it. However, the present invention can be embodied in various different forms and is not limited to the embodiments described herein. In the drawings, in order to clearly explain the present invention, parts not related to the explanation are omitted, and similar reference numerals are given to similar parts throughout the specification.

[0029] Throughout the specification, when it is said that a part is "connected" to another part, this includes not only the case where it is "directly connected", but also the case where it is "electrically connected" with other elements interposed therebetween. Further, when it is said that a part "includes" a certain component, this means that, unless otherwise stated to the contrary, it does not exclude other components and can further include other components.

[0030] As used herein, "server" and "system" refer to a computer configured to include one or more memories (not shown), one or more computer processors (not shown), and one or more programs (not shown). Here, one or more programs (hereinafter, "preprocessing programs") are stored in the memory and configured to be executed by one or more processors. One or more memories, one or more computer processors, and one or more programs may be physically located in the same device and directly connected or connected via a communication network.

[0031] As used herein, an image may mean multi-dimensional data composed of discrete image elements (e.g., pixels in a 2D image and voxels in a 3D image). For example, an image can include medical images obtained by a medical imaging device such as a magnetic resonance imaging (MRI) device, a computed tomography (CT) device, an ultrasonic imaging device, or an X-ray imaging device.

[0032] As used herein, "image restoration" may mean improving the resolution of a low-resolution image, improving the SNR of an image, reducing the aliasing pattern or artifact of an image, or improving the quality of a low-quality image. Also, in the case of MRI, "image restoration" may mean not only the above-mentioned meaning but also processing an image generated from subsampled k-space data to be identical or similar to an image generated from fully sampled k-space data.

[0033] Hereinafter, a medical image restoration system according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

[0034] FIG. 1 is a diagram for explaining the configuration of a medical video restoration and transmission system according to an embodiment of the present invention. Referring to FIG. 1, a medical video restoration and transmission system according to an embodiment of the present invention relates to a medical video restoration and transmission system that communicates with a medical video storage and communication system (PACS: Picture Archiving and Communication System) used in a medical institution to transmit and receive medical video data, and utilizes an artificial neural network model to restore medical video data.

[0035] A medical video restoration and transmission system according to an embodiment of the present invention can be embodied in the form of a cloud computing system (Cloud Computing System). Cloud computing is a computing environment in which IT-related services such as data storage, network, and content usage can be comprehensively used via servers on the Internet. Different from this, the medical video restoration and transmission system can also be embodied by various forms of computing systems that can execute medical video restoration and transmission methods such as server computing, edge computing, and serverless computing.

[0036] A medical video restoration and transmission system according to an embodiment of the present invention can include a communication module, a memory, and a processor.

[0037] The communication module can provide a communication interface to the medical video restoration and transmission system in conjunction with a communication network, and can play a role in transmitting and receiving data with a client terminal, a PACS terminal, and a PACS server described later. Here, the communication module can be a device including hardware and software necessary for transmitting and receiving signals such as control signals or data signals through wired or wireless connection with other network devices.

[0038] On the one hand, in the present invention, the "terminal" can be a wireless communication device with guaranteed portability and mobility, and can be, for example, all types of handheld-based wireless communication devices such as smartphones, tablet PCs, or notebook PCs. Also, the "terminal" can be a wearable device such as a watch, glasses, a hair band, and a ring that has a communication function and a data processing function. Further, the "terminal" can also be a wired communication device such as a PC that can be connected to other terminals or servers via a network.

[0039] The memory can be a recording medium on which a program executed in the medical video restoration and transmission system is recorded. Also, the memory can perform the function of temporarily or permanently storing data processed by the processor. Here, the memory can include volatile storage media or non-volatile storage media, but the scope of the present invention is not limited thereto.

[0040] The processor can control the entire process of the program executed in the medical video restoration and transmission system. Here, the processor can include all types of devices that can process data like a processor. Here, "processor" can mean a data processing device built into hardware that has a physically structured circuit to perform functions represented by, for example, code or instructions included in a program. Thus, as an example of a data processing device built into hardware, it can cover processing devices such as a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a Graphics Processing Unit (GPU), etc., but the scope of the present invention is not limited thereto.

[0041] A medical video restoration and transmission system according to an embodiment of the present invention can include a control server, a manager server, a video restoration server, and an authentication server. When the medical video restoration and transmission system is implemented by a cloud computing system, the medical video restoration and transmission system can include a virtual control server, a manager server, a video restoration server, and an authentication server on the Internet, and these can be included in a single cloud server. For example, in a single cloud server, the virtual control server, manager server, video restoration server, and authentication server have different IP addresses from each other, transmit data based on the different IP addresses from each other, and can perform their respective functions. Alternatively, any of the control server, manager server, video restoration server, and authentication server has the same IP address, and the medical video restoration and transmission method can be executed by data transmission according to a preset algorithm.

[0042] On the other hand, in a medical institution, generally, a client terminal that controls a medical video data shooting device in conjunction with the device or manages medical video data transmission, and a PACS terminal in which a PACS program that allows medical staff to view, process, and manage medical video data can be incorporated may be arranged.

[0043] The client terminal can be a terminal incorporated with a program that provides a user interface (UI) for user login, outputting a work list, and details of video processing. The PACS terminal can be a terminal incorporated with a program that provides a user interface for transmitting medical video data and personal information data stored in a PACS server to the control server of the medical video restoration and transmission system, receiving the medical video data restored by the video restoration server, and storing it in the PACS server.

[0044] The control server of the medical image restoration and transmission system functions to send and receive medical image data with the PACS terminal and the PACS server, and can provide a work list and a video processing information API (Application Programming Interface).

[0045] The manager server receives medical image data from the control server and can automatically distribute video processing tasks to the video restoration server so that the video restoration server can quickly restore the video.

[0046] The video restoration server receives medical image data from the manager server and can perform the received medical image restoration using an artificial neural network model and GPU computing. The artificial neural network model can be a deep learning model pre-trained for video quality improvement processing. The artificial neural network model will be described in detail later.

[0047] The authentication server can provide user authentication and management APIs for the use of the medical image restoration and transmission method program.

[0048] The database can store various data necessary for the medical image restoration and transmission system to execute the program. For example, the database can store a user list, a work list and video processing information, as well as protocol rules and medical image data.

[0049] Figure 2 is a diagram for explaining the network communication between the medical image restoration and transmission system and the PACS system according to an embodiment of the present invention.

[0050] Referring to FIG. 2, the PACS server and the medical video data restoration and transmission system can communicate by utilizing an encrypted wired network based on a site-to-site virtual private network (VPN). In order for the network tunnel to fully protect the data, three elements are required: user authentication, data integrity, and data confidentiality. Such a secure tunnel is called a virtual private network (VPN). In particular, a site-to-site virtual private network can construct an encrypted tunnel between networks in different regions, and can securely share data and resources between the sites via the Internet.

[0051] Specifically, the client terminal can send and receive login information, personal information, and encrypted information of medical video data using the HTTPS protocol by executing authentication server and VPN network communication. HTTPS (Hyper Text Transfer Protocol over Secure Socket Layer) is a version of the HTTP world wide web communication protocol with enhanced security. And the PACS terminal can send and receive DICOM data, which is encrypted medical video data, by executing control server and VPN network communication.

[0052] Hereinafter, a medical video restoration and transmission method using the medical video restoration and transmission system according to an embodiment of the present invention will be described in detail.

[0053] FIG. 3 is a diagram for explaining a medical video restoration and transmission method executed using the medical video restoration and transmission system according to an embodiment of the present invention.

[0054] Referring to FIG. 3, in the medical video restoration and transmission method using the restoration and transmission system according to an embodiment of the present invention, the step (S310) of transmitting the medical video data received by the control server to the manager server can be executed.

[0055] The medical video data can include at least one of the k-space data obtained by accelerated imaging and the DICOM data generated based on the k-space data obtained by accelerated imaging.

[0056] Accelerated imaging, when the medical video data is the video obtained by MRI imaging, can mean reducing the number of repeated imaging times (NEX: Number of Excitations) to shorten the imaging time. For example, it can be to perform the sequence that usually repeats imaging twice once. Also, accelerated imaging can mean obtaining a signal in a narrower range in the phase encoding direction in k-space to obtain an image with low resolution. Also, when the medical video data is the video obtained by MRI imaging, accelerated imaging can mean shortening the MRI imaging time to obtain a subsampled magnetic resonance signal. The subsampled magnetic resonance signal can be a magnetic resonance signal sampled at a sampling rate lower than the Nyquist sampling rate. That is, the accelerated magnetic resonance video can be a video obtained by sampling the magnetic resonance signal at a sampling rate lower than the Nyquist sampling rate. The subsampled magnetic resonance video can be a video including various artificial images, artifacts.

[0057] For example, the number of lines of a fully sampled magnetic resonance signal can be n, and the number of lines of a subsampled magnetic resonance signal can be n / 2. Here, if the degree of reduction of the sampling lines is a multiple of 1 / 2, it can be said that the acceleration index of magnetic resonance imaging is 2. If the degree of reduction of the sampling lines is a multiple of 1 / 3 or 1 / 4, the acceleration indices are 3 and 4, respectively. Also, accelerated imaging can be a method of combining one or more of the above-described imaging methods for imaging.

[0058] The control server can receive medical video data including at least one of the accelerated k-space data and the DICOM data generated based on the accelerated k-space data from a medical video imaging device and a medical video storage and transmission system (PACS).

[0059] The medical video data can include the patient's vital information, medical image information which is a still image of the patient and the treatment site generated at the medical site, and medical video or medical video information taken at the medical site.

[0060] DICOM (Digital Imaging and Communications in Medicine) means a medical digital video and communication standard, and is a general term for various standards used for digital video representation and communication in medical devices.

[0061] DICOM data can mainly include patient information and media characteristics. For example, the diverse medical information data included in DICOM data are patient-related text information collected in the medical field and unprocessed media information, and there are no special restrictions on its format. More specifically, DICOM data can include the biological information of patients, medical image information which is still images of patients and the treated parts generated in the medical field, and medical videos or medical video information taken in the medical field.

[0062] Figures 4a, 4b, and 4c are diagrams for explaining a method for encrypting medical video data according to an embodiment of the present invention.

[0063] Referring to Figures 4a, 4b, and 4c, the control server can encrypt the medical video data before transmitting it to the manager server. Encrypting the medical video data can include shuffling a plurality of video slices included in the medical video data with each other.

[0064] Here, "encryption" can include anonymization. Specifically, anonymization can be to delete all or part of the included personal identification information or convert it into an unknowable form. As anonymization techniques, there are various methods including pseudonym, generalization, permutation, perturbation, and the like.

[0065] For example, a medical video can be a 3D video formed by accumulating a plurality of video slices. Also, such a 3D medical video can be sliced in each direction to generate a plurality of video slices. Shuffling the video slices can be arbitrarily mixing and arranging the video slices arranged in order. Since the 3D medical video can grasp the face and head of the patient and recognize which patient the video belongs to, there is a risk that personal information may leak. To prevent such a risk, the medical video restoration and transmission system shuffles the medical video transmitted from the medical institution and performs anonymization processing to encrypt the personal information by blind processing or deleting metadata such as personal information included in the medical video data, replacing the personal information with a pseudonym.

[0066] For example, a 3D video image of a patient's head can be formed by stacking respective video slices in the coronal plane, sagittal plane, and axial plane. The coronal plane is a cross-section of the head cut in a vertical direction parallel to the face when looking at the face from the front, and these can be stacked on each other to form a 3D video of the head. And the anonymization processing of the 3D video can be executed by changing the order in which the coronal plane slices are stacked and encrypting (see Fig. 4a). The sagittal plane can be a cross-section of the head cut in a vertical direction parallel to the side when looking at the head from the side. The sagittal plane video slices can be stacked on each other to form a 3D video of the head. And the anonymization processing of the 3D video can be executed by changing the order in which the sagittal plane slices are stacked and encrypting (see Fig. 4b). The axial plane is a cross-section that divides the head into upper and lower parts, and these are stacked on each other to form a 3D video of the head. The anonymization processing of the 3D video can be executed by changing the stacking order of each slice and encrypting (see Fig. 4c).

[0067] The control server can store the encrypted medical video data and delete the unencrypted medical video data to prevent personal information leakage. When the control server receives a request for video restoration, it can decrypt the pre-stored encrypted medical video data and transmit the decrypted medical video data to the manager server. The control server can classify the medical video data according to the characteristic information of the medical video data and transmit it to the manager server. The characteristic information can include at least one of the protocol information and sequence information of the medical video data.

[0068] The protocol information is information about various combinations of sequences designed to optimally evaluate videos according to the imaging site or lesion in medical video imaging. The sequence information is, in the case of MRI, a specific setting of the pulse sequence and pulse field gradient, and is information for creating a specific image shape. For example, multi-parametric MRI can include a combination of two or more sequences and / or other special MRI configurations such as spectrometers.

[0069] When the characteristic information includes all of the protocol information and sequence information, the control server can generate a protocol video set based on the received medical video data according to the protocol information. Then, the control server can classify the multiple medical videos included in the protocol video set as sequence information, generate a sequence video set based on the classified sequence information, and transmit the generated sequence video set to the manager server.

[0070] After that, the manager server can execute the step (S320) of transmitting the received medical video data to the video restoration server.

[0071] The manager server can transmit medical video data classified by characteristic information to a plurality of artificial neural networks of a video restoration server pre-classified based on the characteristic information so as to distribute and input it. For example, the manager server can transmit the above-described sequence video set to each artificial neural network model stored in the video restoration server.

[0072] Subsequently, the stage (S330) of restoring the medical video data received by the video restoration server by the artificial neural network model can be executed.

[0073] The artificial neural network model can be a set of algorithms that uses statistical machine learning results to learn the correlation between at least one sub-sampled magnetic resonance image and at least one full-sampled magnetic resonance image. The artificial neural network model can include at least one neural network. The neural network can include network models of methods such as DNN (Deep Neural Network), RNN (Recurrent Neural Network), BRDNN (Bidirectional Recurrent Deep Neural Network), MLP (Multilayer Perceptron), and CNN (Convolutional Neural Network), but is not limited thereto.

[0074] For example, an artificial neural network model can be a model constructed by learning, in units of pixels of at least one sampling line stacked in the phase encoding direction, the correlation between at least one sub-sampled magnetic resonance image and at least one full-sampled magnetic resonance image using a neural network. Also, the artificial neural network model can be constructed using various additional data in addition to the sub-sampled magnetic resonance image and the full-sampled magnetic resonance image. For example, as the additional data, at least one of k-space data corresponding to the magnetic resonance image, real number image data, imaginary number image data, magnitude image data, phase image data, sensitivity data of a multi-channel RF coil, and noise pattern image data can be used.

[0075] FIG. 5 is a diagram for explaining that the artificial neural network model applied according to the characteristic information of medical video data changes according to an embodiment of the present invention. Referring to FIG. 5, when the artificial neural network model restores MRI DICOM medical video data, as an example, the protocol information of the medical video data can be Brain Routine-DN, and this protocol can include sequences such as T2 TSE, T2 FLAIR, and T1 FLAIR. The accelerated medical video imaging by this sequence has the characteristic that the noise in the resulting video is amplified. Therefore, as the artificial neural network model matching such a sequence, one specialized in reducing noise can be applied.

[0076] Also, the protocol can be BrainRoutine-SR, and this protocol can include sequences such as T1SE. The accelerated medical video imaging by this sequence has the characteristic that the resolution of the resulting video decreases. Therefore, as the artificial neural network model matching such a sequence, one specialized in preventing resolution reduction can be applied.

[0077] The protocol can be BrainRoutine-TOF, and this protocol can include sequences such as ToF (Time of Flight). This sequence is a 3D imaging sequence, which can utilize 3D slice direction information. As a result of accelerated imaging, additional post-processing work is required. Therefore, an artificial neural network model that matches such a sequence can be applied to read the 3D slice direction information and perform post-processing work on 3D data.

[0078] Subsequently, the step of transmitting the restored video data to the manager server (S340) can be executed. For example, the manager server can receive a plurality of medical video images formed by a protocol set and / or a sequence set.

[0079] Subsequently, the step of transmitting the restored video data received by the manager server to the control server (S350) can be executed. Subsequently, the step of the control server transmitting the restored video data based on the address where the control server received the medical video data (S360) can be executed. For example, when the control server receives medical video data from the PACS server, the restored video data can be transmitted to the PACS server. The PACS server can encrypt the received restored video data. Thereby, since the PACS server and the medical video restoration and transmission system transmit and receive medical video data using a single tunnel, a strong security communication system for preventing the leakage of patient personal information and the like can be provided.

[0080] The medical video restoration and transmission method according to an embodiment of the present invention described above pseudonymizes and encrypts medical video data including personal information transmitted and received between a medical institution and a medical video restoration and transmission system, providing improved security, so that the leakage of personal information can be prevented.

[0081] According to an embodiment of the present invention, a medical video restoration and transmission method restores a medical video using an optimal artificial neural network model based on the characteristic information of medical video data, so that a medical video with improved quality can be obtained.

[0082] On the other hand, the medical video restoration and transmission method according to an embodiment of the present invention can also be embodied in the form of a recording medium including computer-executable instructions such as program modules executed by a computer. The computer-readable medium can be any available medium accessible by a computer, including all volatile and non-volatile media, as well as removable and non-removable media. Also, the computer-readable medium can include a computer storage medium. The computer storage medium includes all volatile and non-volatile, removable and non-removable media embodied by any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Although the method and system of the present invention have been described in relation to specific embodiments, some or all of its components or operations can be implemented using a computer system having a general-purpose hardware architecture.

[0083] The above description is only an exemplary explanation of the technical idea of the present invention. Those with ordinary knowledge in the technical field to which the present invention pertains can make various modifications and variations without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are not for limiting the technical idea of the present invention, but for explanation purposes, and the scope of the technical idea of the present invention is not limited by such embodiments. The protection scope of the present invention should be interpreted according to the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the rights of the present invention.

Claims

1. A medical image restoration and transmission method using a medical image restoration and transmission system, comprising: receiving medical image data including k-space data and DICOM data generated based on the k-space data; encrypting the medical image data; reconstructing the received medical image data using an artificial neural network model; and transmitting the restored medical image data based on an address of a server that transmitted the medical image data. encrypting the medical image data, obtaining metadata including personal information from the DICOM data, and slicing a 3D image included in the medical image data to obtain a plurality of image slices; encrypting the plurality of video slices by a first means, and encrypting the metadata by a second means different from the first means; Restoring the received medical image data includes: A method for medical image restoration and transmission, comprising: identifying protocol information and sequence information in characteristic information; and identifying an artificial neural network model corresponding to the protocol information and the sequence information from among a plurality of artificial neural networks.

2. 2. The medical image restoration and transmission method of claim 1, wherein the encrypting step comprises shuffling the image slices included in the medical image data relative to one another based on the first means.

3. The medical image restoration and transmission method of claim 1 , wherein the encrypted medical image data is stored and the unencrypted medical image data is deleted.

4. The medical image restoration and transmission method of claim 3 , further comprising: decrypting the stored encrypted medical image data; and restoring the decrypted medical image data using the artificial neural network model.

5. classifying the medical image data according to characteristic information of the medical image data and transmitting the classified medical image data to the artificial neural network model; The medical image restoration and transmission method of claim 1 , further comprising the steps of: distributing and transmitting the medical image data classified according to the characteristic information to a plurality of artificial neural network models pre-classified based on the characteristic information.

6. When the characteristic information includes both the protocol information and the sequence information, The medical image restoration and transmission method of claim 5, further comprising: generating a protocol image set from the medical image data based on the protocol information; classifying a plurality of medical images included in the protocol image set as the sequence information; generating a sequence image set based on the classified sequence information; and transmitting the sequence image set to the artificial neural network model.

7. A medical image restoration and transmission system, comprising: The medical image data receiving device is configured to receive medical image data including k-space data and DICOM data generated based on the k-space data, encrypt the medical image data, restore the received medical image data using an artificial neural network model, and transmit the restored medical image data based on an address where the medical image data is received; The medical image restoration and transmission system is further configured to obtain metadata including personal information from the DICOM data, obtain a plurality of image slices by slicing a 3D image included in the medical image data, encrypt the plurality of image slices by a first means, and encrypt the metadata by a second means different from the first means; The medical image restoration and transmission system is further configured to identify protocol information and sequence information in the characteristic information, and to identify the artificial neural network model corresponding to the protocol information and the sequence information from among a plurality of artificial neural networks.

8. 8. The medical image restoration and transmission system of claim 7, wherein said encrypting step comprises shuffling said plurality of image slices included in said medical image data relative to one another based on said first means.

9. 8. The medical image restoration and transmission system of claim 7, wherein the encrypted medical image data is stored, and the unencrypted medical image data is deleted.

10. The medical image restoration and transmission system of claim 9 , further comprising: decrypting the stored encrypted medical image data; and restoring the decrypted medical image data by the artificial neural network model.

11. classifying the medical image data according to characteristic information of the medical image data and transmitting the classified medical image data to the artificial neural network model; The medical image restoration and transmission system of claim 7, wherein the medical image data classified according to the characteristic information is distributed and transmitted to be input to a plurality of artificial neural network models pre-classified based on the characteristic information.

12. If the characteristic information includes all of the protocol information and sequence information, The medical image restoration and transmission system of claim 7, further comprising: generating a protocol image set from the medical image data based on the protocol information; classifying a plurality of medical images included in the protocol image set as the sequence information; generating a sequence image set based on the classified sequence information; and transmitting the sequence image set to the artificial neural network model.

Images