WO2023239168A1

WO2023239168A1 - Apparatus, method, and computer program for analyzing radiation dose in real time

Info

Publication number: WO2023239168A1
Application number: PCT/KR2023/007819
Authority: WO
Inventors: 한영이; 이동연
Original assignee: 사회복지법인 삼성생명공익재단
Priority date: 2022-06-08
Filing date: 2023-06-08
Publication date: 2023-12-14
Also published as: KR20230168828A

Abstract

According to embodiments of the present disclosure, disclosed is an apparatus for analyzing a radiation dose in real time, comprising: a measuring unit including a scintillator that generates a flashing signal when radiation emitted from a radiation source is irradiated; a recording unit that generates a captured image obtained by photographing the flashing signal generated by the measuring unit, outputs, as an extracted flashing signal, a portion of the captured image where radiation irradiation is performed, by applying a mask generated by using a mask extraction algorithm, and generates and stores a compensation signal obtained by correcting noise generated due to a difference between a pulse repetition frequency of a radiation emitting apparatus and a sampling frequency of a photographing apparatus, by decomposing the flashing light to correspond to a frequency domain through Fourier transformation; and an analyzing unit for analyzing a real-time dose distribution on the basis of the compensation signal.

Description

Real-time radiation dose analysis devices, methods, and computer programs

Embodiments of the present disclosure relate to a real-time radiation dose analysis device, method, and computer program. More specifically, generating a mask corresponding to a captured image using a mask extractor learned with an image synthesis network-based learning model; It is characterized by using a mask to generate a signal for the irradiated part of the image.

Modern radiation therapy consists of protecting normal organs and delivering prescribed doses to malignant tumors through irradiation fields that are modulated over a short period of time. Treatment plans for modern radiation therapy may include three-dimensional conformal therapy, intensity-modulated radiation therapy, stereotactic body radiation therapy, image-guided radiation therapy, respiratory-linked radiation therapy, and cyberknife. Stereoscopic therapy can be applied to tumors occurring in any area and is currently the most basic form of radiation treatment. Intensity-modulated radiation therapy is a cutting-edge radiation treatment method that delivers radiation distribution optimized for the treatment purpose to the treatment area while minimizing radiation irradiation to surrounding normal tissues. According to the intensity-controlled radiation treatment plan, it is important to ensure that radiation is irradiated only to the area to be irradiated.

Therefore, in the case of treatment with intensity-controlled radiation therapy, it is necessary to confirm or predict in advance that the radiation therapy device irradiates radiation to the area where cancer occurs or to the area that requires radiation, but does not irradiate to other areas, before treatment is performed. There is.

Embodiments disclosed herein present a real-time radiation dose analysis device, method, and computer program that captures light generated from radiation irradiated to a scintillator with an imaging device such as a camera and measures dose distribution through the captured images. There is a purpose to doing this.

In addition, embodiments disclosed herein include a real-time radiation dose analysis device, method, and computer program that exists independently of the radiation treatment device and can record and analyze changes in radiation dose over time based on captured images. The purpose is to present .

Devices according to embodiments of the present disclosure include a measuring unit including a scintillator that is irradiated with radiation and emits visible light; A captured image is generated by capturing the visible light emitted from the measuring unit, and the irradiated portion of the captured image is extracted by applying a mask created using a mask extraction algorithm, and output as a scintillation signal through Fourier transform. a recording unit that decomposes the scintillation signal into frequency space to generate and store a compensation signal that corrects noise caused by the difference between the pulse repetition frequency of the radiation emitting device and the sampling frequency of the imaging device; and an analysis unit that analyzes real-time dose distribution based on the compensation signal.

The mask extraction algorithm may be a mask generated using a learning model based on an image synthesis network.

The mask extraction algorithm may be generated by learning from a training data set that uses a captured image as an input and a mask corresponding to the captured image as an output.

The mask extraction algorithm is implemented by including a first generator that converts a captured image into a mask, a second generator that converts the mask into a captured image, and a determination unit that determines whether the mask or the captured image is a real image or a virtual image, and the training data It can be learned and created as a set.

The recording unit extracts a radiation exposure area of the captured image by applying a mask generated using a mask extraction algorithm, calculates an average value of the radiation exposure area, and outputs a scintillation signal of the radiation exposure area based on the average value. can do.

The recording unit converts the scintillation signal into Fourier transform, decomposes it into the imaginary part of the scintillation signal and the real part of the scintillation signal, corrects noise based on the size of the real part of the scintillation signal, and performs the inverse Fourier transform through the imaginary part and the real part. A compensation signal can be extracted through.

The analysis unit may calculate a scintillation operator by applying a luminance calculation formula to the compensation signal and calculate a real-time dose distribution based on the scintillation operator.

The analysis unit may analyze the geometric distortion caused by the position of the imaging device using an iterative reconstruction technique to re-correct the compensation signal, and then calculate a real-time dose distribution from the corrected compensation signal.

The method according to embodiments of the present disclosure includes the steps of generating a flash by irradiating a scintillator of the measuring unit with radiation emitted from a radiation source; generating a captured image by a recording unit capturing the flash of light generated from the measuring unit; The recording unit applying a mask generated using a mask extraction algorithm to extract the irradiated portion of the captured image and output the extracted scintillation signal; The recording unit decomposes the scintillation signal into frequency space through Fourier transform to generate and store a compensation signal that corrects noise caused by a difference between the pulse repetition frequency of the radiation emitting device and the sampling frequency of the imaging device; and an analysis unit receiving the compensation signal and outputting a real-time dose distribution image based on the compensation signal.

The step of outputting the scintillation signal includes extracting a radiation exposure area of the captured image by applying a mask generated using the mask extraction algorithm, calculating an average value of the radiation exposure area, and applying radiation based on the average value. A flash signal in the area can be output.

The step of outputting the scintillation signal includes converting the scintillation signal into Fourier transform, decomposing the scintillation signal into the imaginary part of the scintillation signal and the real part of the scintillation signal, correcting noise based on the size of the real part of the scintillation signal, and dividing the imaginary part and the real part into The compensation signal can be extracted through inverse Fourier transform.

In the step of analyzing the dose distribution, a scintillation operator may be calculated by applying a luminance calculation formula to the compensation signal, and a real-time dose distribution may be calculated based on the scintillation operator.

In the step of analyzing the dose distribution, the geometric distortion caused by the position of the imaging device is analyzed using an iterative reconstruction technique to re-correct the compensation signal, and then a real-time dose distribution can be calculated from the corrected compensation signal.

A computer program according to an embodiment of the present invention may be stored in a medium to execute any one of the methods according to an embodiment of the present invention using a computer.

In addition to this, another method for implementing the present invention, another system, and a computer-readable recording medium for recording a computer program for executing the method are further provided.

Other aspects, features and advantages in addition to those described above will become apparent from the following drawings, claims and detailed description of the invention.

According to one of the above-mentioned problem solving means, a real-time radiation dose analysis device, method, and computer program for photographing light generated by radiation irradiated to a scintillator with an imaging device such as a camera and measuring dose distribution through the photographed image. can be presented.

In addition, it is possible to present a real-time radiation dose analysis device, method, and computer program that exists independently of the radiation treatment device and can record and analyze changes in radiation dose over time based on captured images.

Figure 1 is a configuration diagram schematically showing the configuration of a real-time radiation dose analysis device according to an embodiment of the present invention, and Figure 2 shows the actual design form of the real-time radiation dose analysis device according to an embodiment of the present invention. It is a blueprint.

Figure 3 is a flowchart of a real-time radiation dose analysis method according to an embodiment of the present invention.

Figure 4 is a block diagram of a recording unit of a real-time radiation dose analysis device according to an embodiment of the present invention.

Figure 5 is a block diagram for explaining the analysis unit of the real-time radiation dose analysis device according to an embodiment of the present invention.

Figure 6 is a flowchart of a real-time radiation dose analysis method according to embodiments of the present invention.

FIG. 7A is a diagram illustrating the operation of an image synthesis network-based learning device according to embodiments of the present invention.

FIG. 7B is a diagram illustrating the operation of a mask extractor according to embodiments of the present invention.

Figure 7c is an example diagram of a captured image, a mask, and an image resulting from applying a mask to a captured image, according to embodiments of the present invention.

Figure 7d is a diagram showing the difference in output factor between a captured image and a reference.

Figure 8 is a flowchart of the operation of the signal correction unit according to embodiments of the present disclosure.

Figure 9 is a flowchart of a geometric distortion correction method according to embodiments of the present disclosure.

Figure 10 is a schematic diagram for explaining the generation of a scintillation operator in an image correction unit according to an embodiment of the present invention.

Hereinafter, the configuration and operation of the present invention will be described in detail with reference to embodiments of the present invention shown in the attached drawings.

Since the present invention can be modified in various ways and can have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. The effects and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various forms.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. When describing with reference to the drawings, identical or corresponding components will be assigned the same reference numerals and redundant description thereof will be omitted. .

In this specification, terms such as "learning" and "learning" are not intended to refer to mental operations such as human educational activities, but are terms that refer to performing machine learning through procedural computing. interpret.

In the following embodiments, terms such as first and second are used not in a limiting sense but for the purpose of distinguishing one component from another component.

In the following examples, singular terms include plural terms unless the context clearly dictates otherwise.

In the following embodiments, terms such as include or have mean that the features or components described in the specification exist, and do not exclude in advance the possibility of adding one or more other features or components.

In the drawings, the sizes of components may be exaggerated or reduced for convenience of explanation. For example, the size and thickness of each component shown in the drawings are shown arbitrarily for convenience of explanation, so the present invention is not necessarily limited to what is shown.

In cases where an embodiment can be implemented differently, a specific process sequence may be performed differently from the described sequence. For example, two processes described in succession may be performed substantially at the same time, or may be performed in an order opposite to that in which they are described.

Hereinafter, a real-time radiation dose analysis device according to an embodiment of the present invention will be described with reference to FIGS. 1 and 2. Figure 1 is a configuration diagram schematically showing the configuration of a real-time radiation dose analysis device according to an embodiment of the present invention, and Figure 2 shows the actual design form of the real-time radiation dose analysis device according to an embodiment of the present invention. It is a blueprint.

The real-time radiation dose analysis device 10 may include a measuring unit 100, a recording unit 200, and an analysis unit 300.

The measuring unit 100 is a component that measures an area to which radiation is irradiated, and may include a scintillation plate 110 that emits visible light by irradiating radiation emitted from a radiation source. Radiation can be emitted from a treatment device in a defined area and intensity. The scintillator 100 refers to a phosphor that is excited by emitted radiation particles and emits visible light. The scintillation plate 110 has the advantage of being reusable because it generates visible light by stimulation of radiation, unlike a photosensitive film that is sensitive to radiation. In the present invention, an analysis unit 300 is used to record an image of visible light generated in the recording unit 200, analyze the recorded image, and output an image of the real-time radiation dose, so that the radiation dose is measured in real time at the same time as the measurement. analysis is possible.

The scintillation plate 110 may be placed at the top of the dark room 22, but the arrangement of the scintillation plate 110 can be designed in various ways within the range of measuring real-time dose through recording of visible light.

The recording unit 200 can record a captured image of visible light using visible light generated from the measuring unit 100. The recording unit 200 can generate a captured image by capturing visible light generated from the measuring unit 100, and extract a bright portion from the captured image as the irradiated area. More specifically, the recording unit 200 may apply a mask to the captured image to extract the area or portion where radiation was irradiated.

The recording unit 200 may include a CMOS camera module 21 and a dark room 22 as a device configuration. The CMOS camera module 21 may be provided in two pieces so that they are symmetrically disposed on both sides of the flash plate 110 disposed at the top of the dark room 22. However, the number and arrangement of the CMOS camera module 21 are not limited to this, and in addition to the CMOS camera module 21, it may be another imaging device that can easily record a flash signal. The CMOS camera module 21 may be installed in the dark room 22. The dark room 22 may support the CMOS camera module 21.

The recording unit 200 can record an image captured by visible light generated from the scintillation plate 110 in the form of a video using the CMOS camera module 21. The recording unit 200 may extract the irradiated portion or area from the captured image, perform Fourier transformation on the captured image, and perform correction to remove distortion caused by the frequency difference. The recording unit 200 may perform a method of processing a captured image according to the flowchart of FIG. 6. The processor configuration of the recording unit 200 will be described in detail in the related drawings described later.

The analysis unit 300 may analyze the correction signal received from the recording unit 200 and output a real-time radiation dose distribution image. The analysis unit 300 can generate an image of the dose distribution of radiation irradiated in real time by calculating a scintillation operator by applying a luminance calculation formula to the compensation signal. The operation of the analysis unit 300 will be described in detail in the related drawings described later.

Although not shown in this drawing, the recording unit 200 and the analysis unit 300 may each include a communication unit, a processor, and a memory.

The communication unit can communicate with various types of external devices and servers according to various types of communication methods. Each communication unit of the recording unit 200 and the analysis unit 300 is connected by a network (not shown) and can exchange data with each other.

The processor may perform an operation to overall control each

component

200 and 300 having each memory using various programs stored in the memory. Processors include microprocessors, central processing units (CPUs), processor cores, multiprocessors, application-specific integrated circuits (ASICs), and field programmable gate arrays (FPGAs). It may include a processing device, but the present invention is not limited thereto.

The memory can temporarily or permanently store all types of data processed by each component (200, 300) having each memory. Memory may include random access memory (RAM), read only memory (ROM), and non-permanent mass storage devices such as disk drives, but the present invention is not limited thereto. The operations performed by each of the above-described recording unit 200 and analysis unit 300 may be performed by each processor while communicating with a communication unit of another configuration.

In this drawing, the recording unit 200 and the analysis unit 300 are shown as being provided as separate modules, but they are functionally divided according to the operation performed by the real-time radiation dose analysis device 10 and must be separated independently from each other. This does not mean that it can be implemented by integrating it into one module, if necessary. Of course, depending on the embodiment, the real-time radiation dose analysis device 10 may include other configurations not shown in this drawing.

In S110, the measuring unit 100 of the real-time radiation dose analysis device is irradiated with radiation and emits visible light.

In S120, the recording unit 200 of the real-time radiation dose analysis device can record a captured image of visible light.

In S130, the recording unit 200 of the real-time radiation dose analysis device may apply a mask to extract the irradiated portion from the captured image and convert it into a scintillation signal including the irradiated portion.

In S140, the recording unit 200 of the real-time radiation dose analysis device may decompose the scintillation signal into frequency space through Fourier transform to generate and store a compensation signal that corrects distortion caused by the frequency difference.

In S150, the analysis unit 300 of the real-time radiation dose analysis device may calculate a scintillation operator by applying a luminance calculation formula to the compensation signal and output a real-time dose distribution image based on the scintillation operator.

The recording unit 200 may include a mask extraction unit 210, a signal extraction unit 220, and a signal correction unit 230. When a radiation therapy device is operated according to an intensity-controlled radiation treatment plan, the recorder 200 may obtain a captured image and determine whether the irradiated areas match the radiation irradiated areas planned in the treatment plan.

The mask extractor 210 may obtain a mask image related to the captured image by inputting a captured image of emitted visible light to the mask extractor. The mask extractor 210 may extract a mask by applying a threshold to the cumulative image for each segment when radiation is emitted in an intensity-controlled radiation treatment plan. At this time, the threshold may be a value set based on the brightness value of each pixel. A segment refers to a section where radiation is emitted, and can refer to an area in a captured image where the brightness value is above the threshold. The mask extractor 210 may modulate the irradiation surface in real time with a predetermined control point when radiation is emitted through a three-dimensional intensity controlled rotational radiation treatment plan. In this case, an image of visible light emitted by irradiation may be an image of a modulated irradiation surface.

The mask extractor (MSH, see FIG. 7b) may be generated using an image synthesis network-based learning device.

A learning device based on an image synthesis network may have a structure as shown in FIG. 7A. The image synthesis network-based learning device is to be learned with a training data set of a captured image (TD1, see FIG. 7a) that captures visible light emitted by irradiation and a mask (TD2, see FIG. 7a) corresponding to the captured image. You can. The image synthesis network-based learning device can be generated by learning a first generator (GA1), which has a function of generating a mask from a captured image, and a second generator (GA2), which has a function of generating a captured image from a mask. . The image synthesis network-based learning device can learn a discriminator (DT) that has the function of determining whether a captured image or mask is a real image or a virtual image. Here, the captured images or masks TD1 and TD2 may be videos or frames included in the images. The first generator (GA1), the second generator (GA2), and the discriminator (DT) may be implemented as software, but are not limited to this and may be implemented as hardware. The image synthesis network-based learning device can transmit the first generator (GA1), which generates a mask from a captured image, to the recording unit 200 of the real-time radiation dose analysis device 10 through a mask extractor (MSH, see FIG. 7B). An image synthesis network-based learning device can learn a mask extractor (MSH) by applying a binarization image extraction operator. A learning device based on an image synthesis network can be trained to output an image corresponding to a mask using each frame of a captured video as input. Unlike video, single-frame images can have low image intensity and contain a lot of background noise. Therefore, it may be impossible to extract a mask by applying a threshold value set based on brightness to each frame of the image.

The recording unit 200 may input a captured image (ID) into a mask extractor (MSH) generated by a learning device based on an image synthesis network and extract a mask (OD) corresponding to the captured image.

As shown in FIG. 7B, the mask extractor (MSH) generates a mask (OD) corresponding to the captured image using different methods for the image during intensity-controlled radiation therapy and the image during stereoscopic intensity-controlled rotational radiation therapy. can do. The mask (OD) corresponding to the captured image refers to an image or image to which an effect is applied only to the irradiated area of the captured image.

The signal extraction unit 220 may apply the mask obtained through the mask extraction unit 210 to convert the captured image into a scintillation signal of the irradiated portion.

The signal correction unit 230 may decompose the flash signal into frequency space through Fourier transform and generate and store a compensation signal that corrects distortion caused by the frequency difference. The signal correction unit 230 may generate a compensation signal using the method of FIG. 8.

Figure 5 is a block diagram for explaining the analysis unit 300 of the real-time radiation dose analysis device according to an embodiment of the present invention.

The analysis unit 300 may include an image correction unit 310 and an output unit 320.

The image correction unit 310 may correct geometric distortion caused by the location of a photographing device, for example, a camera. The image correction unit 310 utilized an iterative reconstruction algorithm used in reconstruction of computed tomography.

As shown in FIG. 9, in the image correction unit 310, the captured image may be marked as 'b' in the iterative reconstruction algorithm described later.

The image correction unit 310 may correct geometric distortion using a scintillation operator (A _Lumi ) instead of a projection operator calculated based on the transmission length through which a radiation photon passes through the subject. The scintillation operator (A _Lumi ) can be calculated as a luminance calculation formula for the scintillation signal. As shown in FIG. 10, the luminance (B) of the flash signal is determined by the distance (r) between the recording unit 200 (here, this may mean the CMOS camera module 21) and the subject 120 and the recording unit 200. It can be defined by the brightness of a flash signal generated from an arbitrary area ( _L Hereinafter, the subject 120 may refer to the flash plate 110 described above.

[Equation 1]

The scintillation operator (A _Lumi ) uses the luminance calculation formula of [Equation 1] described above as It can be defined as a determinant calculated for every ₁ to L _x ). The scintillation operator (A _Lumi ) defined in this way can be expressed as [Equation 2] below.

[Equation 2]

At this time, r _N,x in the denominator of [Equation 2] represents the distance from the Nth CMOS camera module 21 to a random area (L _x ) of the subject 120, and L _N,x in the numerator is N It represents the brightness of a flash signal generated from an arbitrary area (L _x ) of the subject 120 detected by the th CMOS camera module 21.

Thereafter, the image correction unit 310 may map the glare operator (A _Lumi ) calculated in this way to each area corresponding to an arbitrary area (L _x ) of the subject 120 of the captured image (b) (x ₀ =A _Lumi ^T ×b). In other words, the initial predicted value (x ₀ ) can be calculated by multiplying each component of the scintillation operator (A _Lumi ) matrix with each region corresponding to an arbitrary region (L _x ) of the subject 120 of the captured image (b). .

In S210, the real-time radiation dose analysis device 10 records the captured image of the scintillation plate.

In S220, the real-time radiation dose analysis device 10 generates a signal including the irradiated portion from the captured image. Specifically, the real-time radiation dose analysis device 10 can be implemented to distinguish between irradiated portions and non-irradiated portions in a captured image and process only the irradiated portion.

In S231, the real-time radiation dose analysis device 10 may extract a mask by processing an image of radiation irradiated by an intensity-controlled radiation treatment plan on a segment basis. In S232, the real-time radiation dose analysis device 10 may extract a mask by processing an image of radiation irradiated by a three-dimensional intensity control rotation radiation treatment plan based on control points.

In S240, the real-time radiation dose analysis device 10 applies a mask to correct the distortion that occurs based on the difference between the frequency of the radiation treatment device and the sample frequency when imaging the imaging device, and analyzes the extracted signal in the frequency dimension (for example, For example, by converting to Fourier transform), distortion occurring through the real part, imaginary part, and DC component of the signal can be selected and the signal with the selected distortion can be output. The real-time radiation dose analysis device 10 can output a compensation signal by applying an inverse Fourier transform after undergoing an iterative attenuation process for the signal with selected distortion. Radiation is irradiated at a fixed dose rate in a radiation therapy device, and the imaging device may interfere with the sampling frequency, resulting in aliasing artifacts.

In S250, the real-time radiation dose analysis device 10 may extract a scintillation signal including the irradiated area by applying the mask extracted in S231 and/or S232. As shown in FIG. 7C, an image 73 including the irradiated area may be output by applying the extracted mask 72 to the captured image 71 (see FIG. 7C).

Additionally, the real-time radiation dose analysis device 10 can correct the output ratio of the scintillation signal by applying a mask. The real-time radiation dose analysis device 10 can calculate the area of the area where radiation is irradiated through the mask and correct the output ratio based on the area. As shown in FIG. 7D, the real-time radiation dose analysis device 10 compares the actual radiation output (reference, output factor) and the radiation output (camera, output factor) by the image captured by the imaging device and takes the image. The radiation output from the image (camera) can be corrected to the actual radiation output (reference). The real-time radiation dose analysis device 10 may image a light intensity that is different from the actual radiation intensity depending on the sensitivity (ISO) of the imaging device.

In S260, the real-time radiation dose analysis device 10 may correct geometric distortion included in the scintillation signal. The real-time radiation dose analysis device 10 can correct geometric distortion caused by, for example, camera position.

The output ratios between the image captured by the imaging device of the recording unit (camera) and the actual emitted radiation (reference) may differ as shown in FIG. 7D. Radiation may scatter from one another to increase the dose, but imaging devices only capture images according to brightness, and differences may occur in the actual imaging ratio (output factor). In the present invention, the output factor of the captured image can be compensated based on this difference.

Figure 8 is a flowchart of the operation of the signal correction unit 230 according to embodiments of the present disclosure.

In S310, the signal correction unit 230 receives a signal extracted by applying a mask from the signal extraction unit 220.

In S320, the signal correction unit 230 may Fourier transform the signal into the frequency dimension. The signal correction unit 230 can decompose the Fourier transform result into the real part (Real(FT(x))) and the imaginary part (Imag(FT(x))) of the signal. In addition, the signal correction unit 230 The signal correction unit 230 can obtain the real part (Real(FT(x)) and the imaginary part (Imag(FT(x)) of the signal, and the DC component (FT(xo)). )) can be used to select the signal (xk). The selected signal (

) can be calculated by the equation below.

In S340, the signal correction unit 230 may attenuate the magnitude of the selected signal. Specifically, the magnitude of the selected signal may be attenuated as shown in the equation below, but is not limited to this and may be attenuated to various magnitudes.

In S341, the signal correction unit 230 may determine whether the processed signal satisfies a predetermined condition through processes such as signal selection and signal attenuation. Here, the predetermined conditions may be as follows, but are not limited to this and various modifications are possible.

In S350, if the signal satisfies a predetermined condition, the signal correction unit 230 may extract a signal that satisfies the condition.

In S360, the signal correction unit 230 converts a signal that satisfies the conditions into an inverse Fourier transform (

) to create a compensation signal with distortion corrected (

) can be output.

As steps S310 to S360 are performed, a compensation signal in which distortion due to the frequency difference included in the signal is compensated may be output. The real-time radiation dose analysis device according to embodiments of the present disclosure may perform steps S310 to S360 for radiation irradiation according to an intensity-controlled radiation treatment plan, but is not limited thereto.

The real-time radiation dose analysis device can correct geometric distortion using a scintillation operator (A _Lumi ) instead of a projection operator calculated based on the transmission length through which radiation photons pass through the subject.

In S410, the real-time radiation dose analysis device 10 may acquire a captured image (b). In S415, the real-time radiation dose analysis device 10 may calculate the scintillation operator (A _Lumi ) as shown in FIG. 10.

refers to the brightness of the subject corresponding to camera N,

refers to the distance between the recording unit corresponding to camera N and the subject.

In S420, the real-time radiation dose analysis device 10 provides an initial predicted value (A Lumi) based on the captured image and the scintillation operator (A _Lumi ).

) can be calculated.

The real-time radiation dose analysis device 10 may repeatedly perform steps S431 to S435.

In S431, the real-time radiation dose analysis device 10 provides the initial predicted value (

) can be input into the objective function to calculate the output value. In S432, the real-time radiation dose analysis device 10 can calculate the minimum value of the objective function.

In S433, the real-time radiation dose analysis device 10 moves during the moving step (

) can be calculated.

In S434, the real-time radiation dose analysis device 10 moves to the moving stage (

) can be considered to determine convergence. In S435, the real-time radiation dose analysis device 10 may calculate a correction image if the convergence is not converged.

In S440, the real-time radiation dose analysis device 10 may output the final image as a real-time radiation distribution image if convergence is determined as a result of convergence.

The device described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), etc. , may be implemented using one or more general-purpose or special-purpose computers, such as a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of 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 optical media (magneto-optical media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with limited examples and drawings, various modifications and variations can be made by those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, other implementations, other embodiments, and equivalents of the claims also fall within the scope of the claims described below.

Claims

A measuring unit including a scintillator that is irradiated with radiation and emits visible light;

Generating a captured image of visible light emitted from the measuring unit,

A mask generated using a mask extraction algorithm is applied to extract the irradiated portion of the image and output it as a scintillation signal,

a recording unit that decomposes the scintillation signal into frequency space through Fourier transform to generate and store a compensation signal that corrects noise caused by the difference between the pulse repetition frequency of the radiation emitting device and the sampling frequency of the imaging device; and

A real-time radiation dose analysis device comprising: an analysis unit that analyzes real-time dose distribution based on the compensation signal.
According to paragraph 1,

The mask extraction algorithm is,

A real-time radiation dose analysis device that is a mask generated using a learning model based on an image synthesis network.
According to paragraph 2,

The mask extraction algorithm is,

A real-time radiation dose analysis device that is learned and created using a training data set that uses captured images as input and masks corresponding to the captured images as output.
According to paragraph 3,

The mask extraction algorithm is,

It is implemented including a first generator that converts the captured image into a mask, a second generator that converts the mask into the captured image, and a determination unit that determines whether the mask or the captured image is a real image or a virtual image, and is learned and generated using the training data set. , Real-time radiation dose analysis device.
According to paragraph 1,

The register is,

Real-time radiation, which extracts the radiation exposure area of the captured image by applying a mask generated using a mask extraction algorithm, calculates an average value of the radiation exposure area, and outputs a scintillation signal of the radiation exposure area based on the average value. Dose analysis device.
According to paragraph 1,

The register is,

The scintillation signal is converted to Fourier transform and decomposed into the imaginary part of the scintillation signal and the real part of the scintillation signal, noise is corrected based on the size of the real part of the scintillation signal, and a compensation signal is generated through inverse Fourier transform through the imaginary part and the real part. Extracting, real-time radiation dose analysis device.
According to paragraph 1,

The analysis unit,

A real-time radiation dose analysis device that calculates a scintillation operator by applying a luminance calculation formula to the compensation signal and calculates a real-time dose distribution based on the scintillation operator.
According to paragraph 1,

The analysis department,

A real-time radiation dose analysis device that analyzes the geometric distortion caused by the position of the imaging device using an iterative reconstruction technique, re-corrects the compensation signal, and then calculates a real-time dose distribution from the corrected compensation signal.
In the real-time dose analysis method of the real-time dose analysis device,

A step of generating a flash by irradiating the scintillator of the measurement unit with radiation emitted from a radiation source;

generating a captured image by a recording unit capturing the flash of light generated from the measuring unit;

The recording unit applying a mask generated using a mask extraction algorithm to extract the irradiated portion of the captured image and output the extracted scintillation signal;

The recording unit decomposes the scintillation signal into frequency space through Fourier transform to generate and store a compensation signal that corrects noise caused by a difference between the pulse repetition frequency of the radiation emitting device and the sampling frequency of the imaging device; and

A real-time radiation dose analysis method comprising: an analysis unit receiving the compensation signal and outputting a real-time dose distribution image based on the compensation signal.
According to clause 9,

The mask extraction algorithm is,

A real-time radiation dose analysis method using a mask generated using an image synthesis network-based learning model.
According to clause 10,

The mask extraction algorithm is,

A real-time radiation dose analysis method that is learned and created using a training data set that uses a captured image as input and a mask corresponding to the captured image as an output.
According to clause 11,

The mask extraction algorithm is,

It is implemented including a first generator that converts the captured image into a mask, a second generator that converts the mask into the captured image, and a determination unit that determines whether the mask or the captured image is a real image or a virtual image, and is learned and generated using the training data set. , real-time radiation dose analysis method.
According to clause 9,

The step of outputting the flash signal is,

Applying a mask generated using the mask extraction algorithm to extract a radiation exposure area of the captured image, calculating an average value of the radiation exposure area, and outputting a scintillation signal of the radiation exposure area based on the average value, in real time. Radiation dose analysis method.
According to clause 9,

The step of outputting the flash signal is,

The scintillation signal is converted to Fourier transform and decomposed into the imaginary part of the scintillation signal and the real part of the scintillation signal, noise is corrected based on the size of the real part of the scintillation signal, and a compensation signal is generated through inverse Fourier transform through the imaginary part and the real part. Extracting, real-time radiation dose analysis method.
According to clause 9,

The step of analyzing the dose distribution is,

A real-time radiation dose analysis method that calculates a scintillation operator by applying a luminance calculation formula to the compensation signal and calculates a real-time dose distribution based on the scintillation operator.
According to clause 9,

The step of analyzing the dose distribution is,

A real-time radiation dose analysis method in which the geometric distortion caused by the position of the imaging device is analyzed using an iterative reconstruction technique to re-correct the compensation signal, and then a real-time dose distribution is calculated from the corrected compensation signal.
A computer program stored in a computer-readable storage medium to execute the method of claim 9 using a computer.