JP2021153952A - Accuracy verification device, radiation treatment system, accuracy verification method and computer program - Google Patents

Abstract

To enable verifying calculation accuracy of an affected part position every patient every perspective angle.SOLUTION: An accuracy verification device 10 is applied to a radiation treatment system 100, acquires a patient's transparent image obtained by photographing a patient 2 by the X-ray fluoroscopy, and calculates a position of a tracking target of a moving body tracking device 60 from the patient's transparent image. The accuracy verification device creates a synthetic transparent image by synthesizing a model transparent image obtained by photographing a tracking target model simulating a tracking target by the X-ray fluoroscopy and the patient's transparent image, and calculates the tracking position accuracy for the tracking target of the moving body tracking device 60 on the basis of the synthetic transparent image.SELECTED DRAWING: Figure 1

Description

The present invention relates to an accuracy verification device, a radiotherapy system, an accuracy verification method and a computer program.

Radiation therapy that irradiates the affected area with various types of radiation such as particle beams such as X-rays and protons is widely used. In radiation therapy, it is necessary to accurately irradiate the affected area. When the position and shape of the affected area change due to internal movements such as breathing and heartbeat, it is required to control the radiation to be irradiated according to those changes. As a method of irradiating radiation with high accuracy, there is moving body tracking radiotherapy.

In moving body tracking radiotherapy, the position of the affected area is measured from a fluoroscopic image of the patient taken by using X-rays or the like. By controlling the irradiation position and irradiation timing of the radiation according to the measured position of the affected area, it is possible to accurately irradiate the affected area in the patient's body.

As a method of measuring the position of the affected part from the fluoroscopic image, there is a method of tracking a metal marker placed in the patient's body. Since the metal marker has a higher density than the human tissue and appears as a high-contrast object in the fluoroscopic image, it can be tracked by a method such as a template matching method.
Further, in recent years, a method called markerless tracking has been proposed in which the position of the affected area is calculated from a fluoroscopic image without using a metal marker.

As a general technique in this technical field, there is a technique described in Patent Document 1.

Patent Document 1 states, "The conversion parameter CP and the inverse conversion parameter RCP used when deriving the target position in the fluoroscopic image TI by learning the target position derived at the time of treatment planning, and the likelihood calculation parameter used for deriving the likelihood in the fluoroscopic image. Using the learning device 110 for deriving the LP, the fluoroscopic image TI acquired when the treatment beam B is irradiated, and various parameters derived by the learning device 110, the target position is used using the likelihood in the fluoroscopic image TI acquired when the treatment beam B is irradiated. With the moving body tracking device 120 that derives the above, the position of the target can be tracked from the fluoroscopic image TI of the patient P under irradiation at high speed and with high accuracy in the radiotherapy. "

JP-A-2019-180799

In Patent Document 1, the likelihood calculation parameters are calculated by machine learning from a DRR image (Digitally Reconstructed Radiograph) that is a virtual projection of a CT image of a patient, and the likelihood (the position where the affected part exists) is said to be calculated from a fluoroscopic image of an actual patient. The technique for calculating the certainty) is described.

However, the technique disclosed in Patent Document 1 has the following problems.

That is, the technique described in Patent Document 1 has a problem that there is no method for verifying the calculation accuracy of the affected portion position for each patient and each perspective angle.

Since the body shape and the position of the affected area differ from patient to patient, the contrast of the human body structure reflected in the fluoroscopic image and the amount of noise in the image vary from patient to patient. In addition, depending on the perspective angle at which the patient is seen through, the human body structure that does not contribute to the calculation of the likelihood may be mainly reflected in the fluoroscopic image, and the position of the affected area may not be calculated correctly. Further, the DRR image is a virtual projection of the CT image of the patient based on the physical model, and the image quality is different from the actual perspective image due to the limitation of the resolution of the CT image and the incompleteness of the physical model. Therefore, it is considered that the parameters learned using the DRR image are close to the optimum values for the DRR image, but may not be optimal for the fluoroscopic image of the actual patient. For this reason, the calculation accuracy of the affected area position can change for each patient and each perspective angle, but there has been no method for knowing the amount of change in the calculation accuracy.

The present invention has been made in view of the above problems, and provides an accuracy verification device, a radiotherapy system, an accuracy verification method, and a computer program capable of verifying the calculation accuracy of the affected portion position for each patient and each perspective angle. There is.

In order to solve the above problems, an accuracy verification device according to one aspect of the present invention is applied to a radiotherapy system, acquires a patient perspective image obtained by photographing a patient by X-ray fluoroscopy, and determines the position of a tracking target from the patient perspective image. It is an accuracy verification device that calculates, and a composite perspective image is created by synthesizing a model perspective image obtained by X-ray fluoroscopy of a tracking target model simulating a tracking target and a patient perspective image, and the composite perspective image is used. Based on this, the tracking position accuracy of the tracking target of the moving object tracking device is calculated.

According to the present invention, it is possible to verify the calculation accuracy of the affected portion position for each patient and each perspective angle.

It is a figure which shows the whole outline structure of the radiotherapy system which concerns on embodiment. It is a figure which shows the schematic functional structure of the accuracy verification apparatus which concerns on embodiment. It is a figure which shows an example of the affected part model created in the modeling part of the accuracy verification apparatus which concerns on Example. It is a figure which shows the outline of the method of making a composite fluoroscopic image by the accuracy verification apparatus which concerns on embodiment. It is a figure which shows an example of the composite fluoroscopic image and correct answer data created by the synthesis part of the accuracy verification apparatus which concerns on Example. It is a figure which shows an example of the processing flow of the accuracy verification apparatus which concerns on embodiment. It is a figure which shows an example of the processing flow of the moving body tracking radiotherapy by the radiotherapy system which concerns on embodiment. It is a figure which shows an example of the experiment using the accuracy verification apparatus which concerns on embodiment. It is a figure which shows an example of the experimental result using the accuracy verification apparatus which concerns on embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. The following description and drawings are examples for explaining the present invention, and are appropriately omitted and simplified for the sake of clarification of the description. The present invention can also be implemented in various other forms. Unless otherwise specified, each component may be singular or plural.

In the figure for explaining the embodiment, the same reference numerals are given to the parts having the same function, and the repeated description thereof will be omitted.

The position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc., in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the position, size, shape, range and the like disclosed in the drawings.

When there are a plurality of components having the same or similar functions, they may be described by adding different subscripts to the same reference numerals. However, when it is not necessary to distinguish between these plurality of components, the subscripts may be omitted for explanation.

The radiation therapy system according to the embodiment will be described with reference to FIGS. 1 to 9.

First, the overall configuration of the radiation therapy system 100 will be described with reference to FIG. FIG. 1 is a diagram showing an overall schematic configuration of a radiotherapy system according to the present embodiment.

In FIG. 1, the radiation therapy system 100 is a device for irradiating the affected portion 3 in the patient 2 with a particle beam composed of a heavy particle beam such as a carbon beam or a proton beam. The radiotherapy system 100 includes an accelerator 101, a beam transport device 102, a treatment table 1 capable of positioning the patient 2, and a radiation irradiation device 7 that irradiates the affected portion 3 in the patient 2 with particle beams supplied from the accelerator 101. ing.

The particle beam is accelerated to the required energy by the accelerator 101, and then guided to the radiation irradiation device 7 by the beam transport device 102. The accelerator 101 can be a synchrotron type accelerator, a cyclotron type accelerator, or various other accelerators.

The radiation irradiation device 7 includes two pairs of scanning electromagnets, a dose monitor, and a position monitor (all of which are not shown). The two pairs of scanning electromagnets are installed in a direction perpendicular to each other, and can deflect the particle beam so that the particle beam reaches a desired position in the plane perpendicular to the beam axis at the position of the affected portion 3. The dose monitor measures the amount of irradiated particle beams. The position monitor detects the position where the particle beam has passed. The particle beam that has passed through the radiation irradiation device 7 reaches the affected area 3.

The particle beam irradiation method is not particularly limited, and a raster scanning method or a line scanning method can be used in which the dose distributions formed by the fine particle beams are arranged and the dose distribution is irradiated according to the shape of the affected area 3.

In addition to the scanning method as described above, an irradiation method such as a wobbler method or a double scatterer method is used to expand the distribution of particle beams and then use a collimator or a bolus to form a dose distribution that matches the shape of the affected area 3. The present invention can also be applied to the above.

When X-rays are used instead of particle beams such as carbon beams and proton beams as the radiation used for treatment, an X-ray irradiation device that generates X-rays is provided instead of the accelerator 101 and the beam transport device 102. When γ-rays are used, a γ-ray irradiation device that generates γ-rays is provided instead of the accelerator 101 and the beam transport device 102. When an electron beam is used, an electron beam irradiation device for generating an electron beam is provided instead of the accelerator 101 and the beam transport device 102.

The radiotherapy system 100 includes a CT imaging device 20, a treatment planning device 30, and a patient database (DB) 40 as components involved in formulating a treatment plan.

The CT imaging device 20 captures a CT image of the patient 2 in advance for formulating a treatment plan. The treatment planning device 30 formulates a treatment plan based on the imaging result of the CT imaging device 20. The patient DB 40 stores information about patient 2. In this embodiment, an example using a CT image taken by the CT imaging device 20 is shown, but the MRI image captured by the MRI (Magnetic Resonance Imaging) imaging device or the ultrasonic imaging device was used. Other three-dimensional images taken by another imaging device, such as an ultrasonic image, may be used.

The radiotherapy system 100 includes X-ray imaging devices 6A and 6B, a moving body tracking device 60, and an accuracy verification device 10 as components related to the position measurement of the affected area 3.

The X-ray imaging devices 6A and 6B capture an X-ray fluoroscopic image of patient 2. The moving body tracking device 60 acquires an X-ray fluoroscopic image, performs image processing based on the tracking parameters created in advance, and calculates the position of the affected area 3. The accuracy verification device 10 verifies the calculation accuracy of the position of the affected portion 3 calculated by the moving object tracking device 60. Details of the accuracy verification device 10 will be described later.

The X-ray imaging devices 6A and 6B are a two-dimensional dose of X-ray generators 5A and 5B that generate X-rays toward the patient 2 and X-rays generated from the X-ray generators 5A and 5B and transmitted through the patient 2. It includes two-dimensional X-ray detectors 4A and 4B for detecting the distribution, and a signal processing circuit (not shown).

For the two-dimensional X-ray detectors 4A and 4B, for example, a flat panel detector (FPD) or an image intensifier (I.I .: Image Intensifier) is used. The two-dimensional X-ray detectors 4A and 4B may be any two-dimensional detector other than the FPD or I.I., as long as they acquire a two-dimensional image.

In the present embodiment, the number of X-ray imaging devices is two, but the number is not limited to this. The number of X-ray imaging devices may be any number as long as it is one or more. When calculating the position of the affected area 3 as a three-dimensional position with respect to the treatment room coordinate system, two or more X-ray imaging devices may be used.

As the image processing performed by the moving object tracking device 60, for example, there is a template matching method. In this case, the moving object tracking device 60 creates in advance a template image in which the area in which the affected portion 3 is shown is cut out from the DRR image or the fluoroscopic image as a tracking parameter.

The moving body tracking device 60 compares the X-ray perspective images taken by the X-ray imaging devices 6A and 6B with the template image, and calculates the position having the highest degree of similarity as the position of the affected portion 3 on each X-ray fluoroscopic image. Normalized cross-correlation (NCC) or the like is used as the degree of similarity. After that, the moving body tracking device 60 calculates the position of the affected part 3 as a three-dimensional position with respect to the treatment room coordinate system from the position of the affected part 3 on each X-ray perspective image using the principle of triangulation.

The moving body tracking device 60 in the present embodiment calculates the position of the affected area 3 as a three-dimensional position with respect to the treatment room coordinate system, but the present invention is not limited to this. The moving body tracking device 60 may calculate only the two-dimensional position on each fluoroscopic image taken by the X-ray imaging devices 6A and 6B as the position of the affected portion 3. Further, the moving body tracking device 60 may calculate the contour of the affected portion 3 reflected on the fluoroscopic image as the position of the affected portion 3. Further, the moving body tracking device 60 may calculate the likelihood indicating the certainty that the affected part 3 is present as the position of the affected part 3.

When calculating the two-dimensional position on the fluoroscopic image as the position of the affected part 3, the moving object tracking device 60 may perform image processing using, for example, the above-mentioned template matching method. In this case, the moving object tracking device 60 creates a template image in advance as a tracking parameter.

Further, when calculating the contour of the affected part 3 reflected on the perspective X-ray image as the position of the affected part 3, the moving body tracking device 60 identifies by machine learning based on, for example, a learning image including the affected part 3 and a teacher image including the affected part features. Image processing may be performed in which a device is created and an X-ray perspective image is input to the created classifier to calculate the contour of the affected area 3. In this case, the moving object tracking device 60 creates a parameter of the classifier as a tracking parameter in advance by machine learning. As the machine learning method here, a neural network, a convolutional neural network, an SVM (Support Vector Machine), a decision tree, or the like can be applied.

Further, when calculating the likelihood indicating the certainty that the affected part 3 exists as the position of the affected part 3, the moving object tracking device 60 performs image processing using, for example, the technique described in Patent Document 1. You may. In this case, the moving object tracking device 60 creates in advance the likelihood calculation parameters, the X-ray fluoroscopic image, and various parameters derived from the learning device as tracking parameters.

The radiotherapy system 100 defines a virtual reference point for aligning the position of the patient 2 on the treatment table 1 with the predetermined position defined in the treatment plan. The reference point may be included in a plane on which fluoroscopy is taken by the radiographers 6A and 6B.

In addition, the radiotherapy system 100 includes a network 50 for communicating and connecting various devices to each other.

The accuracy verification device 10 of the present embodiment will be described in detail with reference to FIG. 1 and FIG. 2 showing a schematic functional configuration of the accuracy verification device 10 according to the embodiment.

The accuracy verification device 10 includes a calculation unit 11, a display unit 12, an input unit 13, a storage unit 14, and a communication unit 15. The calculation unit 11 includes a modeling unit 110, an adjustment unit 111, a synthesis unit 112, and a verification unit 113.

The modeling unit 110 creates a model of the affected area based on the shape of the affected area 3 defined on the CT image taken by the CT imaging device 20, for example. The adjusting unit 111 adjusts the position of a fluoroscopic image (hereinafter, referred to as a model fluoroscopic image) obtained by X-ray fluoroscopy of the affected area model based on the input result of the input unit 13. The synthesizing unit 112 synthesizes the model fluoroscopy image adjusted by the adjusting unit 111 and the fluoroscopic image of the patient 2 taken by X-ray fluoroscopy (hereinafter referred to as a patient fluoroscopy image) to create a composite image. The verification unit 113 calculates the calculation accuracy of the position of the affected portion 3 of the moving object tracking device 60 by using the composite fluoroscopic image created by the composite unit 112.

The display unit 12 displays a patient perspective image, a model perspective image, a composite perspective image, and the like on a display device (not shown). The input unit 13 includes an interface such as a mouse, a keyboard, and a switch, and inputs information to the accuracy verification device 10 according to a user operation. The storage unit 14 stores the synthesis parameters used by the synthesis unit 112, the composite fluoroscopic image created by the synthesis unit 112, and the like. The communication unit 15 communicates with the moving object tracking device 60, the CT imaging device 20, and the like.

The processing executed by each component of the accuracy verification device 10 will be specifically described below.

First, the modeling unit 110 reads out the CT image taken by the CT imaging device 20 and the treatment plan created by the treatment planning apparatus 30 from the patient DB 40. Then, based on the structural information of the affected area designated on the CT image by the treatment plan, a model of the affected area having the same structure as that of the affected area 3 is modeled.

FIG. 3 shows an affected area model 200 created based on a treatment plan of a lung cancer patient and a CT image as an example of the affected area model created by the modeling unit 110.

For example, a 3D printer is used for modeling the affected area model 200. The procedure for creating the affected area model 200 when a 3D printer is used for modeling is shown below. First, based on the structural information of the affected area designated on the CT image by the treatment plan, an affected CT image is created by extracting only the affected area in the CT image. After that, the affected CT image is converted into an input data format (for example, STL format: Stereolithography) of a 3D printer, and the affected area extraction modeling data is created. The finally converted affected part extraction modeling data is input to a 3D printer to create an affected part model 200.

In the present embodiment, a 3D printer is used for modeling the affected area model 200, but the present invention is not limited to this. For example, the affected part model 200 may be created by carving out a resin mass so as to have a structure similar to that of the affected part 3. Further, in the present embodiment, the affected area model 200 is created for each patient, but the present invention is not limited to this. For example, in the affected area model 200, it is possible to reuse several types of affected area models 200 prepared in advance without creating the affected area model 200 for each patient.

Based on the position information input by the user in the input unit 13, the adjustment unit 111 projects the affected part model 200 created by the modeling unit 110 onto the model perspective image by moving in parallel the model perspective image taken by X-ray fluoroscopy. An adjusted model perspective image is created by adjusting the position of the affected area model 200.

The synthesizing unit 112 synthesizes the adjusted model fluoroscopy image created by the adjusting unit 111 and the patient fluoroscopy image of the patient 2 taken by X-ray fluoroscopy to create a composite fluoroscopy image. The composite fluoroscopic image is used to measure the calculation accuracy of the position of the affected portion 3 of the moving object tracking device 60.

FIG. 4 conceptually shows a method of creating a composite fluoroscopic image. In the creation of the composite fluoroscopic image, four fluoroscopic images (perspective images 304, 324, 344, 364) taken in the following four situations (situations 300, 320, 340, 360) are considered.

Situation 300 is a case where there is no subject. The incident X-ray 301 passes through the air and becomes a transmitted X-ray 302. In this case, a clear object is not projected on the perspective image 304 detected by the two-dimensional X-ray detector 4A or 4B, and the two-dimensional intensity distribution of the incident X-ray 301 is directly reflected in the image.

Situation 320 is a case where the patient 2 is present as a subject. The incident X-ray 321 passes through the patient 2 and becomes a transmitted X-ray 322. In this case, the fluoroscopic image 324 detected by the two-dimensional X-ray detector 4A or 4B projects the internal structure of the patient 2 and the affected area projection image 3', which is a projection image of the affected area 3 in the patient 2.

Situation 340 is a case where the affected part model 200 exists as a subject. The incident X-ray 341 passes through the affected area model 200 and becomes a transmitted X-ray 342. In this case, the affected part model projected image 200', which is a projected image of the affected part model 200, is projected on the perspective image 344 detected by the two-dimensional X-ray detector 4A or 4B.

Finally, the situation 360 is a virtually assumed case in which the patient 2 and the affected area model 200 exist as subjects. The incident X-ray 361 passes through the patient 2 and the affected area model 200, and becomes a transmitted X-ray 362. In this case, the perspective image 364 detected by the two-dimensional X-ray detector 4A or 4B includes the internal structure of the patient 2, the affected part projection image 3'which is a projection image of the affected part 3 in the patient 2, and the affected part model. The affected area model projection image 200', which is a projection image of 200, is projected.

The compositing unit 112 calculates the perspective image 364 obtained from the perspective image 304, the perspective image 324, and the perspective image 344 in a situation in which the situation 360 is virtually assumed, based on the following mathematical formula.

Here, I ₀ is the pixel value of the perspective image 304, I ₁ is the pixel value of the perspective image 324, I ₂ is the pixel value of the perspective image 344, μ ₁ is the X-ray attenuation coefficient in the patient 2, and T 1 is. The thickness of patient 2 and μ ₂ represent the X-ray attenuation coefficient in the _{affected area model 200, and T 2} represents the thickness of the affected area model 200. That is, by calculating the product of the line attenuation coefficient and the thickness of each object from each perspective image and estimating the X-ray attenuation using the sum of them, two perspective images of the two objects taken separately are used. Perspective images of objects taken at the same time can be combined.

Further, the fluoroscopic image 364 can be calculated based on the following mathematical formula.

Here, α is a parameter representing the effect that the X-ray attenuation decreases as the object is behind due to the influence of the X-ray beam hardening. The value of α takes 0 to 1, and the input unit 13 determines it based on the user's operation.

Further, the fluoroscopic image 364 can be calculated based on the following mathematical formula.

Here, β is a parameter for correcting the difference in the amount of X-ray attenuation due to the difference in density between the affected part model 200 and the affected part 3 in the patient 2.

For example, assuming that the affected area 3 in the patient 2 is a lung tumor, the CT value of the lung tumor (the pixel value of the lung tumor on the CT image, which is approximately proportional to the density of the object) is around -100. On the other hand, if the affected area model 200 is created using, for example, a general resin material, the CT value of the affected area model 200 is around 100. β is a parameter for correcting such a difference between the CT value of the affected part model 200 and the CT value of the affected part 3 in the patient 2, and even if the CT values of the affected part model 200 and the affected part 3 are different, the affected part model 200 is more. It is possible to calculate the tracking accuracy in a situation where the CT value of is close to that of the affected area 3. β takes an arbitrary value and can be calculated using, for example, the ratio of the CT values of the affected area model 200 and the affected area 3. Alternatively, β may be determined by the input unit 13 based on the user's operation.

Further, a noise effect can be added to the composite perspective image created by the mathematical formula (1), the mathematical formula (4) or the mathematical formula (5) based on the following mathematical formula.

Here, I _{1 + 2} is a composite perspective image calculated by the formula (1), the formula (4), or the formula (5), and I'1 _{+ 2} is a noise composite perspective image to which the noise effect is added, and η. (0, σ ₂ ) is a random variable that follows a Gaussian distribution with mean 0 and variance σ. The value of σ has an arbitrary size, and the input unit 13 determines it based on the operation of the user. By adding a noise effect to the composite fluoroscopic image by the mathematical formula (6), it is possible to calculate the tracking accuracy in a situation where tracking is more difficult.

The correct answer data of the affected area model 200 projected on the composite fluoroscopic image is, for example, using a threshold to separate the fluoroscopic image 344 into two values, that is, the area where the affected area model 200 is projected and the area where the affected area model 200 is not projected. Can be created with. Otsu's binarization method or the like can be applied to the threshold calculation method.

FIG. 5 shows an example of the composite fluoroscopic image and the correct answer data created by the synthesis unit 112.

In the composite fluoroscopic image, the internal structure of the patient 2, the affected part projection image 3'which is a projection image of the affected part 3, and 200', which is a projection image of the affected part model 200, are projected. Further, the correct answer data is binarized into the area 401 on which the affected area model 200 is projected and the area 402 on which the affected area model 200 is not projected. The correct answer data was created as two areas, but it is not limited to this. For example, correct answer data may be created as three areas such as an area on which the affected area model 200 is projected, an area on which the affected area model 200 is not projected, and a boundary area between them. Further, correct answer data may be created as two-dimensional information such as the position of the center of gravity of the area where the affected area model 200 is projected.

First, the verification unit 113 transmits the composite fluoroscopic image created by the synthesis unit 112 to the moving object tracking device 60. The verification unit 113 operates the moving body tracking device 60, and the moving body tracking device 60 tracks the affected part model 200 on the synthetic fluoroscopic image and calculates the two-dimensional position of the affected part model 200 on the synthetic fluoroscopic image. After that, the verification unit 113 calculates a comparison result between the tracking position calculated by the moving object tracking device 60 and the correct answer data created by the synthesis unit 112. As a comparison result, for example, a square error between the two-dimensional position of the affected part model 200 on the synthetic fluoroscopic image calculated by the moving object tracking device 60 and the position of the center of gravity of the region 401 on which the affected part model 200 is projected on the correct answer data, etc. Is used.

The display unit 12 displays various images such as a patient perspective image and a composite perspective image, a comparison result calculated by the verification unit 113, and the like on a display device (not shown). The input unit 13 includes an interface such as a mouse, a keyboard, and a switch, and inputs information to the accuracy verification device 10 according to a user operation. The input unit 13 determines the position information when the model fluoroscopic image is translated by the adjustment unit 111 according to the operation of the user. For example, a two-dimensional position is used as the position information determined by the input unit 13. In this case, there is only one type of adjusted model perspective image created by the adjusting unit 111.

Further, the position information determined by the input unit 13 may be a plurality of two-dimensional positions or the like. In this case, the adjusting unit 111 creates a plurality of types of adjusted model fluoroscopic images in which the positions of the projected affected area model 200 are different, and at the same time, a plurality of types of composite fluoroscopic images created by the composite unit 112 are also created.

The storage unit 14 stores the synthesis parameters used by the synthesis unit 112, the composite fluoroscopic image created by the synthesis unit 112, and the like. The communication unit 15 communicates with the moving object tracking device 60, the CT imaging device 20, and the like.

The accuracy verification device 10, the moving object tracking device 60, and the like may be realized by causing the arithmetic unit 11, which is an arithmetic unit, to execute a program. This arithmetic unit may be a computer provided with a CPU, memory, an interface, or the like, or a programmable arithmetic device such as an FPGA (Field Programmable Gate Array). The program read into the arithmetic unit is stored in the internal recording medium or the external recording medium MM (see FIG. 1), and is read by the arithmetic unit.

Further, the operation commands for the accuracy verification device 10, the moving object tracking device 60, and the like may be combined into one program, divided into a plurality of programs, or a combination thereof. In addition, part or all of the program may be realized by dedicated hardware, or may be modularized. Further, various programs may be installed in each device from the program distribution server, the internal storage medium, or the external storage medium MM. Further, the accuracy verification device 10 and the moving object tracking device 60 and the like do not have to be independent, and two or more of them may be integrated and shared to share only the processing.

A part of the plurality of components included in the accuracy verification device 10 may be configured by an external computer. The external computer may be directly connected to the accuracy verification device 10 or may be connected to a communication line such as the Internet. A part or all of the plurality of components included in the accuracy verification device 10 may be individually configured by an electronic circuit as hardware.

For the storage unit 14 included in the accuracy verification device 10, for example, a RAM, a ROM, a hard disk, a USB memory, an SD card, or the like may be used. The storage unit 14 may be a storage on a communication line such as the Internet. One component included in the accuracy verification device 10 may be composed of a plurality of computers that perform distributed processing.

Next, the details of the processing of the accuracy verification device 10 of the radiotherapy system 100 according to this embodiment will be described. FIG. 6 is a flowchart showing a processing procedure of the accuracy verification device 10.

When the accuracy verification device 10 starts processing, the communication unit 15 reads the CT image and the treatment plan from the patient DB 40 and stores them in the storage unit 14 (S501).

Next, the modeling unit 110 has a structure similar to that of the affected area 3 based on the structural information of the affected area designated on the CT image by the treatment plan from the CT image and the treatment plan stored in the storage unit 14. Model 200 (S502).

After that, the communication unit 15 communicates with the moving object tracking device 60, and sets imaging parameters for photographing the affected area model 200 with the X-ray imaging devices 6A and 6B (S503). The imaging parameters refer to the tube voltage and tube current of the X-ray generators 5A and 5B, the angle of fluoroscopy, and the like.

Subsequently, the communication unit 15 communicates with the moving object tracking device 60, photographs the X-ray fluoroscopy of the affected model 200, acquires the model fluoroscopy image, and stores it in the storage unit 14 (S504). Further, the input unit 13 determines the position information when the model fluoroscopic image is translated by the user's operation (S505). Subsequently, the adjusting unit 111 reads out the model fluoroscopic image from the storage unit 14 and creates an adjusted model fluoroscopic image based on the position information determined in S505 (S506).

In parallel with S502 to S506, the communication unit 15 communicates with the motion tracking device 60 to position the patient 2 on the treatment table 1 (S507). The positioning of the patient 2 may be performed by aligning the position of the patient 2 with a predetermined position determined in the treatment plan based on the marking laser light (not shown) provided in the radiation irradiation device 7.

Next, the communication unit 15 communicates with the moving object tracking device 60, and sets imaging parameters for imaging the affected area model 200 with the X-ray imaging devices 6A and 6B (S508). The imaging parameters refer to the tube voltage and tube current of the X-ray generators 5A and 5B. Subsequently, the communication unit 15 communicates with the moving object tracking device 60, takes an X-ray fluoroscopy of the patient 2, acquires the patient fluoroscopy image, and stores it in the storage unit 14 (S509).

After S506 to S509, the input unit 13 accepts the user's operation as necessary and determines the synthesis parameter (S510). The composite parameter is, for example, the value of α in the mathematical formula (4).

Next, the synthesis unit 112 reads out the adjusted model fluoroscopy image and the patient fluoroscopy image stored in the storage unit 14, creates the composite fluoroscopy image and the correct answer data, and stores them in the storage unit 14 (S511). Subsequently, the communication unit 15 communicates with the moving object tracking device 60, reads out the tracking parameters created in advance (S512), transmits the composite fluoroscopic image stored in the storage unit 14 to the moving object tracking device 60, and displays the composite fluoroscopic image on the composite fluoroscopic image. The projected image of the affected area model 200 is tracked (S513).

After that, the verification unit 113 compares the tracking result calculated by the moving object tracking device 60 with the correct answer data stored in the storage unit 14, and displays the comparison result on the display unit 12 (S514).

Further, the communication unit 15 communicates with the moving object tracking device 60 and determines whether or not to shoot with other shooting parameters (S515). When imaging with other imaging parameters for reasons such as treatment with a plurality of fluoroscopic angles (YES in S515), the process returns to S503 and S508, the imaging parameters are changed, and the process proceeds again.

The details of the processing of the moving body tracking radiotherapy by the radiotherapy system 100 according to this embodiment will be described. FIG. 7 is a flowchart showing a processing procedure of moving body tracking radiotherapy by the radiotherapy system 100. The flowchart shown in FIG. 7 is divided into a pretreatment flow F600 executed before treatment and a treatment flow F650 executed when patient 2 visits the hospital and actually irradiates radiation.

When the radiotherapy system 100 starts processing, the CT imaging device 20 captures a CT image of the patient 2 (S601). Next, the treatment planning device 30 creates a treatment plan based on the CT image and stores it in the patient DB 40 (S602).

Subsequently, the moving body tracking device 60 creates tracking parameters based on the treatment plan and the CT image (S603). Further, the accuracy verification device 10 compares the tracking result of the affected area 3 calculated by the moving body tracking device 60 with the correct answer data, and verifies the tracking accuracy (S604).

After that, the radiotherapy system 100 determines whether the tracking accuracy calculated in S604 is within a predetermined range set in advance (S605), and if it is within the predetermined range (YES in S605), the treatment flow F650 is reached. Proceed with. If it is determined that the range is out of the predetermined range (NO in S605), the process returns to S603, and the moving object tracking device 60 recreates the tracking parameter.

If the tracking accuracy is still not within the predetermined range in S605 even after the tracking accuracy is verified a plurality of times in S604, the affected part 3 of the patient 2 is treated with, for example, a metal marker in the treatment flow F650. You may make a suggestion.

In the treatment flow F650, the radiotherapy system 100 performs positioning of the patient 2 on the treatment table 1 (S651). The positioning of the patient 2 may be performed by aligning the position of the patient 2 with a predetermined position determined in the treatment plan based on the marking laser light (not shown) provided in the radiation irradiation device 7.

Next, the moving object tracking device 60 reads out the tracking parameters created in S603 (S652). Further, the moving body tracking device 60 communicates with the X-ray imaging devices 6A and 6B to acquire a fluoroscopic image of the patient 2 taken by X-ray fluoroscopy (S653).

Subsequently, the moving body tracking device 60 calculates the affected portion position from the fluoroscopic image acquired in S653 (S654), and determines whether the affected portion position is within a predetermined range set in advance (S655). If it is within the predetermined range (YES in S655), the radiotherapy system 100 irradiates the affected portion 3 in the patient 2 with a particle beam (S656). In S655, when the affected part position is out of the predetermined range (NO in S655), the process returns to S653.

Further, the radiation therapy system 100 determines whether the irradiated particle dose has reached a predetermined value (S657). When the irradiated particle dose reaches a predetermined value in S657 (YES in S657), the treatment is terminated. In S657, when the irradiated particle dose does not reach a predetermined value (NO in S657), the process returns to S653.

The operations of the radiotherapy system 100 and the accuracy verification device 10 according to this embodiment will be described below.

FIG. 8 shows an explanatory diagram of an experimental example of the radiotherapy system 100 according to the present embodiment. An experiment was conducted to verify the tracking accuracy of the moving object tracking device 60 using the actual fluoroscopic image shown in FIG. 8 and the three fluoroscopic images of Examples and Comparative Examples. The actual perspective image is an image actually taken by fluoroscopy, and the position of the affected portion 3 is known. An example is an image created by synthesizing a fluoroscopic image of an affected area model 200 in an actual fluoroscopic image using the radiotherapy system 100 of the present embodiment. A comparative example is an image calculated by virtually seeping a CT image through X-rays.

In the actual fluoroscopic image and the comparative example, the affected area 3 projected in the image was tracked by the template matching method, and the tracking accuracy was calculated. In the example, using the radiotherapy system 100 of the present embodiment, the projected image of the affected area model 200 synthesized in the image was tracked by the template matching method, and the tracking accuracy was calculated.

FIG. 9 shows an explanatory diagram of the experimental results of the experimental example. In FIG. 9, the tracking error in the actual fluoroscopic image was 5.8 (pixels) on average and 22.1 (pixels) at the maximum. The tracking error in the examples was 6.0 (pixels) on average and 20.0 (pixels) at the maximum, which was equivalent to the tracking error in the actual fluoroscopic image. The tracking error in the comparative example was 5.2 (pixels) on average and was 14.1 (pixels) at the maximum, which was smaller than the tracking error in the actual fluoroscopic image and the example.

That is, the tracking error in the DRR image tends to be underestimated than the tracking error in the real fluoroscopic image. On the other hand, in the embodiment, the tracking error closer to the actual fluoroscopic image than the DRR image can be calculated. As a result, the tracking error in the actual fluoroscopic image, which could not be calculated in the past, can be calculated by using the composite fluoroscopic image in the present embodiment.

Further, in the present embodiment, by creating the affected area model 200 for each patient, it is possible to evaluate the tracking error of the moving body tracking device 60 that changes for each patient.

Further, in the present embodiment, a composite fluoroscopic image is created for each angle of fluoroscopy of the patient 2, and a tracking error is calculated for each composite fluoroscopic image, so that the tracking error of the moving object tracking device 60 that changes for each fluoroscopic angle is obtained. Can be evaluated.

Further, in the present embodiment, by creating a composite fluoroscopic image after correcting the effect of beam hardening, the contrast of the affected area model 200 can be made more realistic, and the tracking error of the moving object tracking device 60 can be made more realistic. It can be evaluated accurately.

Further, in the present embodiment, the contrast of the affected area model 200 can be made more realistic by creating a synthetic fluoroscopic image after correcting the difference between the CT values of the affected area model 200 and the affected area 3 in the patient 2. , The tracking error of the moving object tracking device 60 can be evaluated more accurately.

Further, in the present embodiment, by adding a noise effect to the composite fluoroscopic image, it is possible to evaluate the tracking accuracy in a situation where tracking is more difficult in a state where the photographing parameters are lowered.
Further, in the present embodiment, by creating the affected area model 200 for each patient, it is possible to create a composite fluoroscopic image that matches the shape of the affected area 3 in the patient's body, which changes for each patient, and the tracking error of the moving body tracking device 60. Can be evaluated more accurately.

The present invention is not limited to the above-described examples, and includes various modifications. For example, the above-described embodiment has been described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to those having all the described configurations. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add / delete / replace a part of the configuration of each embodiment with another configuration.

Further, each of the above configurations, functions, processing units, processing means and the like may be realized by hardware by designing a part or all of them by, for example, an integrated circuit. Further, each of the above configurations, functions, and the like may be realized by software by the processor interpreting and executing a program that realizes each function. Information such as programs, tables, and files that realize each function can be placed in a memory, a recording device such as a hard disk or SSD, or a recording medium such as an IC card, SD card, or DVD.

In addition, the control lines and information lines indicate those that are considered necessary for explanation, and do not necessarily indicate all the control lines and information lines in the product. In practice, it can be considered that almost all configurations are interconnected.

1 ... Treatment table 2 ... Patient 3 ... Affected area 3'... Projection image of the affected area projected on a fluoroscopic image 4A, 4B ... 2D X-ray detector 5A, 5B ... X-ray generator 6A, 6B ... X-ray imaging device 7 ... Particle beam irradiation device 10 ... Accuracy verification device 11 ... Calculation unit 12 ... Display unit 13 ... Input unit 14 ... Storage unit 15 ... Communication unit 20 ... CT imaging device 30 ... Treatment planning device 40 ... Patient DB 50 ... Network 100 ... Radiation therapy system 110 ... Modeling part 111 ... Adjustment part 112 ... Synthesis part 113 ... Verification part 200 ... Affected part model

Claims (11)

It is an accuracy verification device applied to a radiotherapy system, which acquires a patient fluoroscopic image of a patient taken by X-ray fluoroscopy and calculates the position of a tracking target of a moving body tracking device from the patient fluoroscopic image.
A composite perspective image is created by synthesizing a model perspective image obtained by photographing the tracking target model simulating the tracking target by X-ray fluoroscopy and the patient perspective image.
An accuracy verification device for calculating the tracking position accuracy of the tracking target of the moving object tracking device based on the composite fluoroscopic image. The accuracy according to claim 1, wherein the composite fluoroscopic image is calculated using a parameter representing the attenuation of X-rays in the patient's body and a parameter representing the attenuation of X-rays in the model to be tracked. Verification device. To calculate the synthetic fluoroscopic image using a parameter representing the attenuation of X-rays in the patient's body, a parameter representing the attenuation of X-rays in the model to be tracked, and a parameter representing the effect of beam hardening. The accuracy verification device according to claim 1. The synthesis using a parameter representing the attenuation of X-rays in the patient body, a parameter representing the attenuation of X-rays in the tracking target model, and a parameter representing the attenuation of the X-rays of the tracking target in the patient body. The accuracy verification device according to claim 1, wherein a fluoroscopic image is calculated. The accuracy verification device according to claim 1, wherein a noise effect is applied to the composite fluoroscopic image. The accuracy verification device according to claim 1, wherein the tracking position accuracy is calculated for each fluoroscopic parameter. The accuracy verification device according to claim 1, wherein the tracking target model is created for each patient. The accuracy verification device according to claim 1, wherein the tracking target model is created by using a 3D modeling device. It is a radiotherapy system equipped with an accuracy verification device that acquires a patient fluoroscopy image of a patient by X-ray fluoroscopy and calculates the position of a tracking target of a moving object tracking device from the patient fluoroscopy image.
The accuracy verification device is
A composite perspective image is created by synthesizing a model perspective image obtained by photographing the tracking target model simulating the tracking target by X-ray fluoroscopy and the patient perspective image.
A radiotherapy system characterized in that the tracking position accuracy of a tracking target of the moving object tracking device is calculated based on the composite fluoroscopic image. It is an accuracy verification method by an accuracy verification device applied to a radiotherapy system, which acquires a patient fluoroscopy image of a patient by X-ray fluoroscopy and calculates the position of a tracking target of a moving body tracking device from the patient fluoroscopy image.
A composite perspective image is created by synthesizing a model perspective image obtained by photographing the tracking target model simulating the tracking target by X-ray fluoroscopy and the patient perspective image.
An accuracy verification method for calculating the tracking position accuracy of the tracking target of the moving object tracking device based on the composite fluoroscopic image. A computer program applied to a radiotherapy system and executed by a computer that acquires a patient fluoroscopic image of a patient taken by X-ray fluoroscopy and calculates the position of a tracking target of a moving body tracking device from the patient fluoroscopic image.
A function of creating a composite perspective image by synthesizing a model perspective image obtained by photographing a tracking target model simulating the tracking target by X-ray fluoroscopy and the patient perspective image.
A computer program that realizes a function of calculating the tracking position accuracy of the tracking target of the moving object tracking device based on the composite fluoroscopic image.

Images