KR20230051374A - Method and apparatus for providing information to monitor lung disease progression - Google Patents

Abstract

The present disclosure relates to a method of providing information for tracking the progression of lung disease and a device, therefore. The method of providing information for tracking the progress of lung disease according to an embodiment of the present disclosure comprises the steps of: extracting a 3D first lung area from a 3D chest tomography (CT) image through a learned first network function; detecting a lesion area from the first lung area extracted through a learned second network function; generating a 2D image of a second lung area by projecting the first lung area including the detected lesion area onto a coronal plane; extracting a third lung area from a chest radiography (X-ray) image through a learned third network function; transforming the second lung area based on the third lung area and calculating an output value of a loss function for similarity to the third lung area; and matching a second closed area with a third closed area by optimizing the second closed area based on the output value of the loss function. According to the present invention, a process of manually or semi-automatically performing matching for comparison with the image is automated, thereby saving the required manpower and time.

Description

Method for providing information for tracking the progress of lung disease and apparatus therefor

The technical idea of the present disclosure (disclosure) relates to a method for providing information for tracking the progress of lung diseases and an apparatus therefor.

In the diagnosis and treatment of patients with chronic pneumonia, chronic obstructive pulmonary disease (COPD), and idiopathic pulmonary fibrosis (IPF) requiring continuous observation, when determining the worsening or improvement of the disease through imaging, different time A diagnosis is made by reading a number of images obtained in

For example, chest tomography (CT) is generally not used repeatedly due to its high radiation dose, but it can pinpoint the location of disease onset, so it is based on 1-2 tomographic images and has a relatively low radiation dose. Multiple chest X-ray images can be used to determine the course of disease progression.

At this time, since it is inevitable that the location and size of the object shown on the image will change due to the location of the image acquisition device, the setting of the imaging device, and various structures required for the treatment process, in order to more easily observe the disease process, they are set to a standard size and A process of matching each other by location, that is, a matching process is required.

Conventionally, such matching was performed using a manual or semi-automatic method.

Specifically, in manual registration, the user performs geometric transformation (movement, rotation, enlargement and reduction) on the image using an input device and manual registration software, and semi-automatic registration allows the user to make an anatomically meaningful and visually distinguishable images. Three or more easy parts (key points) are equally selected from the matching original (source) image and matching target (target) image, and geometric transformation is performed on the target image to match the original image.

However, this conventional method is dependent on the difference in the shade of the image according to the progression of the disease, the difference in the image according to the structure for treatment (arterial tube, venous tube, or airway intubation, etc.), and the parameter setting (dose, etc.) of the device for image acquisition. Accordingly, there is a problem in that matching errors or errors occur. In addition, in registering multiple images to one original registration, it takes a lot of time and manpower for the user to correct errors caused by semi-automatic or automatic registration, as well as the cost of user training for this problem. also had

A technical task to be achieved by a method for providing information for tracing the progress of lung disease and an apparatus therefor according to the technical idea of the present disclosure is to automatically match a CT image and an X-ray image to improve the quality of patients with chronic lung disease. An object of the present invention is to provide a method capable of providing information for progress tracking and a device therefor.

Technical tasks to be achieved by a method for providing information for tracking the progress of lung diseases and an apparatus therefor according to the technical idea of the present disclosure are not limited to the tasks mentioned above, and other tasks not mentioned will be clear to those skilled in the art from the description below. will be understandable.

According to one aspect of the technical concept of the present disclosure, a method for providing information for tracking the progress of lung diseases includes extracting a first 3-dimensional lung region from a 3-dimensional chest tomography (CT) image through a learned first network function. doing; detecting a lesion area from the first lung area extracted through a learned second network function; generating a two-dimensional image of a second lung region by projecting the first lung region including the detected lesion region onto a coronal plane; extracting a third lung region from a chest X-ray image through a learned third network function; calculating an output value of a loss function for similarity with the third closed region while modifying the second closed region based on the third closed region; and matching the second closed region with the third closed region by optimizing the second closed region based on the output value of the loss function.

According to an exemplary embodiment, the calculating of the output value of the loss function for the similarity with the third lung region may include performing at least one of movement, rotation, enlargement, and reduction of the second lung region, and the loss This can be done by calculating the output value of the function.

According to an exemplary embodiment, the matching with the third closed region may include repeating the step of calculating an output value of a loss function for similarity with the third closed region so that the output value of the loss function reaches a minimum value. can be performed

According to an exemplary embodiment, when the chest tomography image and the chest radiographic image are captured, setting a search range for at least one of movement, rotation, enlargement, and reduction based on a standard protocol for the location of the subject. Further, the calculating of the output value of the loss function for the similarity with the third lung region may be performed by changing at least one of a movement amount, a rotation amount, an enlargement amount, and a reduction amount within the search range.

According to an exemplary embodiment, the method may further include synthesizing the optimized second lung region at a position corresponding to the third lung region of the chest radiographic image.

According to one aspect according to the technical idea of the present disclosure, an information providing device for tracking the progress of a lung disease includes at least one processor; a memory for storing a program executable by the processor; and by executing the program, the processor extracts a 3D first lung region from a 3D chest tomography (CT) image through the learned first network function, and extracts the extracted first lung region through the learned second network function. A lesion area is detected from the first lung area, a 2D image of the second lung area is generated by projecting the first lung area including the detected lesion area onto a coronal plane, and the learned third network function extracts a third lung region from a chest X-ray image, transforms the second lung region based on the third lung region, and outputs a loss function for similarity with the third lung region The second closed region may be matched with the third closed region by calculating and optimizing the second closed region based on the output value of the loss function.

According to the information providing method and apparatus for tracing the progress of lung disease according to embodiments according to the technical idea of the present disclosure, when a user (radiologist) reads an image of a patient with chronic lung disease, Since comparison becomes easy, the time and effort required for reading and diagnosis can be greatly reduced.

According to the information providing method and apparatus for tracing the progress of lung diseases according to embodiments according to the technical idea of the present disclosure, the process of performing manual or semi-automatic registration for comparison with images is automated, requiring manpower and time. can save

According to the information providing method and apparatus for tracking the progress of lung diseases according to embodiments according to the technical idea of the present disclosure, unlike the process of matching the entire existing image, since only the lung region is used as a criterion, the amount of calculation is large. The width can be reduced, and at the same time, the conversion error of the image can be reduced.

Effects obtainable by a method for providing information for tracking the progress of lung diseases and a device therefor according to the technical idea of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned are the present disclosure from the description below. It will be clearly understood by those skilled in the art.

In order to more fully understand the drawings cited in this disclosure, a brief description of each drawing is provided.
1 is a flowchart illustrating a method according to an embodiment of the present disclosure.
FIG. 2 is a flowchart illustrating an embodiment of step S150 of FIG. 1 .
3 to 6 are views for illustratively explaining a process of matching lung regions according to an embodiment of the present disclosure.
7 is a block diagram briefly showing the configuration of a device according to an embodiment of the present disclosure.

Since the technical spirit of the present disclosure may be subject to various changes and may have various embodiments, specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the technical spirit of the present disclosure to specific embodiments, and should be understood to include all changes, equivalents, or substitutes included in the scope of the technical spirit of the present disclosure.

In describing the technical idea of the present disclosure, if it is determined that a detailed description of a related known technology may unnecessarily obscure the subject matter of the present disclosure, the detailed description will be omitted. In addition, numbers (eg, first, second, etc.) used in the description process of the present disclosure are only identifiers for distinguishing one component from another component.

In addition, in the present disclosure, when one component is referred to as "connected" or "connected" to another component, the one component may be directly connected or directly connected to the other component, but in particular Unless otherwise described, it should be understood that they may be connected or connected via another component in the middle.

In addition, terms such as "~ unit", "~ group", "~ character", and "~ module" described in the present disclosure mean a unit that processes at least one function or operation, which includes a processor, a micro Processor (Micro Processor), Micro Controller, CPU (Central Processing Unit), GPU (Graphics Processing Unit), APU (Accelerate Processor Unit), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array) may be implemented by hardware or software or a combination of hardware and software.

In addition, it is intended to make it clear that the classification of components in the present disclosure is merely a classification for each main function in charge of each component. That is, two or more components to be described below may be combined into one component, or one component may be divided into two or more for each more subdivided function. In addition, each component to be described below may additionally perform some or all of the functions of other components in addition to its main function, and some of the main functions of each component may be performed by other components. Of course, it may be dedicated and performed by .

Throughout this specification, network function may be used interchangeably with neural network and/or neural network. Here, a neural network (neural network) may be composed of a set of interconnected computational units, which may be generally referred to as nodes, and these nodes may be referred to as neurons. A neural network is generally composed of a plurality of nodes. Nodes constituting a neural network may be interconnected by one or more links.

Some of the nodes constituting the neural network may form one layer based on distances from the first input node. For example, a set of nodes having a distance of n from the first input node may constitute n layers.

The neural network described herein may include a deep neural network (DNN) including a plurality of hidden layers in addition to an input layer and an output layer.

In addition, the method according to an embodiment of the present disclosure may be performed in a personal computer having computing capability, a workstation, a computer device for a server, or a separate device for this purpose.

Also, the method may be performed on one or more computing devices. For example, at least one or more steps of the method 100 according to an embodiment of the present disclosure may be performed in a client device and other steps may be performed in a server device. In this case, the client device and the server device may be connected through a network to transmit and receive calculation results. Alternatively, method 100 may be performed by distributed computing technology.

Hereinafter, embodiments of the present disclosure will be described in detail in turn.

1 is a flowchart illustrating a method according to an embodiment of the present disclosure.

In step S110, the device may extract (segmentation) a 3D first lung region from a 3D chest tomography (CT) image through the learned first network function.

The chest tomography image may be received from an external database server or may be acquired through imaging from an imaging device connected to the device through wired or wireless communication.

For example, the chest tomography image may be a 3D image generated by continuously imaging a body part including the chest of the subject in one direction through a computed tomography method. Alternatively, according to embodiments, the chest tomography image may be generated by stacking a plurality of 2D slice images generated through continuous imaging.

The first network function may be one in which lung region extraction has been previously learned through learning data (eg, a chest tomography image in which lung regions are labeled through experts, examinations, etc.).

The information about the extracted first lung region may include information about 3-dimensional position coordinates. Here, the location coordinates of the lung region may be generated corresponding to the same reference coordinates as those of the 3D chest tomography image.

In step S120, a 2D image of the second lung region may be generated by projecting the extracted first lung region onto the coronal plane.

In an embodiment, step S120 may be performed by digitally reconstructed radiographic-projection (DRR) projection of the extracted 3-dimensional first lung region on the coronal plane. However, this is an example, and various methods may be applied to generate a 2D image of the second lung region.

Information on the generated second lung region may include information on 2-dimensional position coordinates.

Meanwhile, although not shown, the method 100 may further include, before step S120, detecting a lesion area from the first lung region extracted through step S110. For example, the device may detect the lesion area from the first lung area through the learned second network function.

In this case, the second network function may be learning for detection of the lesion region in advance through learning data (eg, a 3D lung region image in which the lesion region is labeled through an expert, examination, etc.). Information on the detected lesion area may include information on 3-dimensional position coordinates.

In an embodiment, step S120 may be performed on the first lung region including the detected lesion region. In this case, the generated 2D second lung area may include information about the 2D transformed lesion area.

In step S130, the apparatus may extract a third lung region from at least one chest X-ray image through the learned third network function.

The chest radiographic image may be received from an external database server or acquired through imaging from an imaging device connected to the device through wired or wireless communication. Also, the chest radiographic image may include a plurality of images captured at different time points over time.

The third network function may be one in which lung region extraction has been previously learned through learning data (eg, a chest radiographic image in which lung regions are labeled through experts, examinations, etc.).

The extracted information on the third lung region may include information on 2-dimensional position coordinates. Here, the location coordinates of the lung region may be generated corresponding to the same reference coordinates as those of the 2-dimensional chest radiographic image.

Meanwhile, although not shown, the method 100 may further include detecting a lesion area from the extracted third lung area. For example, the device may detect the lesion area from the third lung area through the learned fourth network function. The information on the lesion area detected in the third lung region may be used to determine the progress of the lung disease through a plurality of image comparisons later.

In step S140, the device may set one of the second lung region and the third lung region as a reference image. The reference image refers to an image that is a reference image for matching, and as described below, matching (or optimization) is performed by transforming another image based on the reference image.

In an embodiment, step S140 may be set based on a user's selection input. However, this is exemplary, and the device may be configured to automatically set an image set as a basic value among the second lung region and the third lung region as the reference image.

In step S150, the device may transform the second or third lung region based on the reference image and optimize the second or third lung region to match the reference image.

As shown in FIG. 2 , in an embodiment, step S150 may include steps S151 to S153.

In step S151, the device may set a search range for at least one of movement, rotation, enlargement, and reduction based on a standard protocol for the location of the subject.

That is, since the posture or position of the patient follows the standard protocol in the acquisition process of various medical images, the device moves, rotates, A search range for at least one of enlargement and reduction may be set.

In steps S152 and S153, the device may optimize the second closed region or the third closed region based on the output value of the loss function while performing transformation on the second closed region or the third closed region within the search range. .

Specifically, the device calculates the output value of the loss function for the similarity with the reference image while changing at least one of the movement amount, the rotation amount, the enlargement amount, and the reduction amount of the second lung region or the third lung region within the search range. can do. Subsequently, it is determined whether the output value of the loss function reaches a minimum value, and the process of converting the second closed region or the third closed region and calculating the output value of the loss function may be repeatedly performed until the minimum value is reached.

That is, the loss function outputs a lower value as the modified second lung region or third lung region becomes more similar to the reference image, and an iterative process of reducing the output value of the loss function (transformation -> measuring the output value of the loss function) ), it is possible to perform matching with the reference image by optimizing the second lung region or the third lung region.

To this end, stochastic gradient descent or the like may be applied, but is not limited thereto, and various statistical methods may be applied.

In particular, according to the present disclosure, the amount of computation for optimization can be drastically reduced by limiting the search range so that the second lung region or the third lung region is modified only within a certain range based on a standard protocol.

In step S160, the optimized second lung region or third lung region may be combined with the 2D original image related to the reference image.

In an embodiment, in the case of the third lung region, which is the reference image, the device may synthesize the optimized second lung region at a location corresponding to the third lung region of the chest radiographic image.

In an embodiment, when the reference image is the second lung region, the device generates a chest image generated by converting a 3D chest tomography image into a 2D image, and converts an optimized third lung region into the second lung region of the chest image. It can be synthesized at a location corresponding to the lung region. In this case, generation of the chest image may be performed by digitally reconstructed radiographic-projection (DRR) projection of the 3D chest tomography image on the coronal plane.

According to the present disclosure, it is possible to easily compare the synthesized image generated in step S160 with a plurality of chest radiographic images taken at different time points over time, so that the degree of lung disease progression can be easily read. It happens.

In an embodiment, the generation of the second lung region may be performed based on a chest image generated by projecting a 3D chest tomography image. That is, the device may generate an image of the second lung region by extracting the 2D second lung region from the chest image through the learned network function.

Meanwhile, although not shown, the method 100 compares the first lesion area detected from the first lung area and included in the second lung area with the second lesion area detected from the third lung area to obtain progress information. The step of providing may be further included. For example, the device may be configured to provide information such as the size, location, shape change, etc. of the first lesion area and the second lesion area, and determine the progress of the lung disease based on the corresponding information.

3 to 6 are views for illustratively explaining a process of matching lung regions according to an embodiment of the present disclosure.

First, referring to FIG. 3 , a 3D first lung region may be extracted by inputting a chest tomography image to a learned network function. The chest tomography image is 3D data, and may be generated by, for example, stacking a plurality of slice images.

Subsequently, referring to FIG. 4 , the extracted 3D image of the first lung region may be converted into a 2D image including a second lung region corresponding to the first lung region through DRR projection. By extracting the lung region rather than the entire chest tomography image and performing 2D transformation on it, the amount of computation can be greatly reduced.

Subsequently, referring to FIG. 5 , a third lung region may be extracted from an X-ray image through the learned network function. A chest X-ray image is two-dimensional data and may include a plurality of images captured at different points in time.

Next, referring to FIG. 6 , a synthesized image may be generated by matching the second lung region and the third lung region with each other. For example, optimization (or matching) may be performed by using the second lung region as a reference image and transforming the third lung region until the output value of the loss function for similarity with the reference image reaches a minimum value. there is. Thereafter, the optimized image of the third lung region may be combined with a chest image generated by projecting the chest tomography image.

Meanwhile, in FIG. 6 , a third lung region optimized for a 2-dimensional chest image generated by projecting a chest tomography image using the second lung region as a reference image is illustrated, but as described above, the third lung region is synthesized. Of course, the second lung region optimized for the chest radiographic image may be synthesized by using the region as a reference image and then optimizing the second lung region.

7 is a block diagram briefly showing the configuration of a device according to an embodiment of the present disclosure.

The communication unit 710 may receive input data (such as a CT image of the chest) for providing information for matching lung regions and/or tracking the progress of lung diseases. The communication unit 710 may include a wired/wireless communication unit. When the communication unit 710 includes a wired communication unit, the communication unit 710 may include a local area network (LAN), a wide area network (WAN), a value added network (VAN), and a mobile communication network ( mobile radio communication network), a satellite communication network, and one or more components that enable communication through a mutual combination thereof. In addition, when the communication unit 710 includes a wireless communication unit, the communication unit 710 transmits and receives data or signals wirelessly using cellular communication, a wireless LAN (eg, Wi-Fi), and the like. can In an embodiment, the communication unit may transmit/receive data or signals with an external device or an external server under the control of the processor 740 .

The input unit 720 may receive various user commands through external manipulation. To this end, the input unit 720 may include or connect one or more input devices. For example, the input unit 720 may be connected to an interface for various inputs such as a keypad and a mouse to receive user commands. To this end, the input unit 720 may include an interface such as a thunderbolt as well as a USB port. In addition, the input unit 720 may receive an external user command by including or combining various input devices such as a touch screen and buttons.

The memory 730 may store programs and/or program commands for operation of the processor 740 and may temporarily or permanently store input/output data. The memory 730 is a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), RAM , SRAM, ROM (ROM), EEPROM, PROM, magnetic memory, a magnetic disk, it may include at least one type of storage medium.

In addition, the memory 730 may store various network functions and algorithms, and may store various data, programs (one or more instructions), applications, software, commands, codes, etc. for driving and controlling the device 700. there is.

The processor 740 may control the overall operation of the device 700 . Processor 740 may execute one or more programs stored in memory 730 . The processor 740 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to the technical idea of the present disclosure are performed.

In an embodiment, the processor 740 extracts a 3D first lung region from a 3D chest tomography (CT) image through a learned first network function, and projects the extracted first lung region onto a coronal plane. By doing so, a 2D image of the second lung region is generated, a third lung region is extracted from a chest X-ray image through the learned second network function, and the second lung region and the third lung region are extracted. set one of them as a reference image, transform the second lung region or the third lung region based on the reference image, calculate an output value of a loss function for similarity with the reference image, and calculate The second lung region or the third lung region may be matched with the reference image by optimizing the second lung region or the third lung region based on the output value.

In an embodiment, the processor 740 generates a 2D chest image by projecting the 3D chest tomography image onto a coronal plane, and extracts the second lung region from the chest image through a learned third network function. can

In an embodiment, the processor 740 detects a lesion region from the first lung region extracted through the learned fourth network function, and projects the first lung region including the detected lesion region onto a coronal plane. A 2D image of the second lung region may be generated.

In an embodiment, the processor 740 performs at least one of movement, rotation, enlargement, and reduction of the second lung region or the third lung region, and calculates an output value of a loss function for similarity with the reference image. can do.

In an embodiment, the processor 740 may perform matching with the reference image by repeating the step of calculating the output value of the loss function for the degree of similarity with the reference image so that the output value of the loss function reaches a minimum value. there is.

In an embodiment, the processor 740 sets a search range for at least one of movement, rotation, enlargement, and reduction based on a standard protocol for the location of the subject when the chest tomography image and the chest radiographic image are captured, and , an output value of a loss function for similarity with the reference image may be calculated by changing at least one of a movement amount, a rotation amount, an enlargement amount, and a reduction amount within the search range.

In an embodiment, when the reference image is the third lung region, the processor 740 may synthesize the optimized second lung region at a position corresponding to the third lung region of the chest radiographic image.

In an embodiment, when the reference image is the second lung region, the processor 740 generates a 2D chest image by projecting the 3D chest tomography image onto a coronal plane, and generates the optimized third lung region. It may be synthesized at a location corresponding to the second lung region of the chest image.

The method according to an embodiment of the present disclosure may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the present disclosure, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler.

In addition, the method according to the disclosed embodiments may be provided by being included in a computer program product. Computer program products may be traded between sellers and buyers as commodities.

A computer program product may include a S/W program and a computer-readable storage medium in which the S/W program is stored. For example, a computer program product may include a product in the form of a S/W program (eg, a downloadable app) that is distributed electronically through a manufacturer of an electronic device or an electronic marketplace (eg, Google Play Store, App Store). there is. For electronic distribution, at least a part of the S/W program may be stored in a storage medium or temporarily generated. In this case, the storage medium may be a storage medium of a manufacturer's server, an electronic market server, or a relay server temporarily storing SW programs.

A computer program product may include a storage medium of a server or a storage medium of a client device in a system composed of a server and a client device. Alternatively, if there is a third device (eg, a smart phone) that is communicatively connected to the server or the client device, the computer program product may include a storage medium of the third device. Alternatively, the computer program product may include a S/W program itself transmitted from the server to the client device or the third device or from the third device to the client device.

In this case, one of the server, the client device and the third device may execute the computer program product to perform the method according to the disclosed embodiments. Alternatively, two or more of the server, the client device, and the third device may execute the computer program product to implement the method according to the disclosed embodiments in a distributed manner.

For example, a server (eg, a cloud server or an artificial intelligence server) may execute a computer program product stored in the server to control a client device communicatively connected to the server to perform a method according to the disclosed embodiments.

Although the embodiments have been described in detail above, the scope of the present disclosure is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concepts of the present disclosure defined in the following claims are also within the scope of the present disclosure. belongs to

Claims (7)

In the information providing method for tracking the progress of lung disease,
extracting a 3-dimensional first lung region from a 3-dimensional chest tomography (CT) image through a learned first network function;
detecting a lesion area from the first lung area extracted through a learned second network function;
generating a two-dimensional image of a second lung region by projecting the first lung region including the detected lesion region onto a coronal plane;
extracting a third lung region from a chest X-ray image through a learned third network function;
calculating an output value of a loss function for similarity with the third closed region while modifying the second closed region based on the third closed region; and
matching the second closed region with the third closed region by optimizing the second closed region based on the output of the loss function. According to claim 1,
Calculating the output value of the loss function for the similarity with the third closed region,
The method is performed by performing at least one of movement, rotation, enlargement, and reduction of the second lung region, and calculating an output value of the loss function. According to claim 2,
The step of matching with the third lung region,
The method is performed by repeating the step of calculating the output value of the loss function for the similarity with the third closed region so that the output value of the loss function reaches a minimum value. According to claim 2,
Further comprising setting a search range for at least one of movement, rotation, enlargement, and reduction based on a standard protocol for a location of the subject when the chest tomography image and the chest radiographic image are captured;
Calculating the output value of the loss function for the similarity with the third closed region,
A method performed by changing at least one of a movement amount, a rotation amount, an enlargement amount, and a reduction amount within the search range. According to claim 1,
The method further comprising synthesizing the optimized second lung region at a position corresponding to the third lung region of the chest radiographic image. In the information providing device for tracking the progress of lung disease,
at least one processor;
a memory for storing a program executable by the processor; and
The processor, by executing the program, extracts a 3-dimensional first lung region from a 3-dimensional chest tomography (CT) image through the learned first network function, and extracts the extracted lung region through the learned second network function. A lesion area is detected from the first lung area, a 2D image of the second lung area is generated by projecting the first lung area including the detected lesion area onto a coronal plane, and a learned third network function is generated. A third lung region is extracted from a chest radiograph (X-ray) image, the second lung region is transformed based on the third lung region, and the output value of the loss function for the similarity with the third lung region is and matching the second closed region with the third closed region by optimizing the second closed region based on an output value of the loss function. A computer program stored in a recording medium to execute the method of any one of claims 1 to 5.

Images