Bed position control device, bed position control method, and radiotherapy device

Abstract

The present invention enables more efficient positioning control of a bed on which a patient is placed. A bed position control device that controls a position of a bed on which a patient is placed includes: an image acquisition unit that acquires a fluoroscopic X-ray image obtained by imaging the patient by X-ray fluoroscopy; a pseudo fluoroscopic X-ray image acquisition unit that acquires a pseudo fluoroscopic X-ray image created from a three-dimensional image of the patient; a second region-of-interest creation unit that, from a first region-of-interest set in the pseudo fluoroscopic X-ray image so as to include a region of a target structure, creates a second region-of-interest including the region of the target structure in accordance with a positional deviation of the patient; and an image matching unit that matches an image in the second region-of-interest in the pseudo fluoroscopic X-ray image with an image in the second region-of-interest in the fluoroscopic X-ray image, determines a position of the bed necessary for aligning the position of the patient in the fluoroscopic X-ray image, and moves the bed to the determined position of the bed.

Description

Patient Bed Position Control Device, Patient Bed Position Control Method, and Radiation Therapy Device

[0001] The present invention relates to a patient bed position control device, a patient bed position control method, and a radiation therapy device.

[0002] As one of the cancer treatment methods, there is radiation therapy in which radiation is irradiated to a patient. The radiation used in radiation therapy is roughly classified into uncharged particle beams such as X-rays and gamma rays, and charged particle beams such as proton beams and carbon beams. Treatment using the latter charged particle beam (hereinafter also referred to as a charged particle beam) is generally called particle beam therapy.

[0003] The dose imparted by the uncharged particle beam decreases at a constant rate from a shallow position to a deep position in the body. On the other hand, the charged particle beam can form a dose distribution (black curve) having an energy peak at a specific depth. Therefore, by adjusting the peak to the position of the tumor, it is possible to significantly reduce the dose irradiated to normal tissues located deeper than the tumor.

[0004] In radiation therapy, irradiating the target tumor with a charged particle beam of a desired dose as accurately as possible leads to an improvement in the treatment effect. In order to realize accurate irradiation of the charged particle beam to the tumor, it is necessary to create a treatment plan in a treatment planning device in advance and align the patient at the same position as in the treatment plan during actual irradiation. This alignment of the patient is called patient positioning.

[0005] As a method for patient positioning in radiation therapy, there is a method using a fluoroscopic X-ray image (Digital Radiography: DR) obtained by photographing a patient from two directions with two sets of X-ray tubes and flat detectors arranged orthogonally to each other. A projection processing image (hereinafter referred to as a pseudo-fluoroscopic X-ray image) created by arranging the fluoroscopic X-ray image of the patient and the CT (Computed Tomography) image at the time of treatment planning in a virtual space and calculating the attenuation amount of X-rays entering from the virtual X-ray tube to the virtual flat detector is compared, and the displacement amount required to match the positions of the patient on both images is obtained.

[0006] The displacement is determined so that a structural object serving as a positioning marker on the fluoroscopic X-ray image (hereinafter referred to as the positioning target structure) coincides with the same location on the simulated fluoroscopic X-ray image. Generally, bone is used as the structural object serving as the marker. The displacement is determined by manual visual alignment by a medical professional (hereinafter also referred to as the operator) on the image matching device, or by automatic alignment using an automatic calculation function.

[0007] Generally, fluoroscopic X-ray images may include structures other than the target structure for positioning, such as the patient's immobilization devices or soft tissues. Alternatively, even the bones, which are the target structure, may change in their arrangement from day to day. In such situations, the fluoroscopic X-ray image and the pseudo-fluoroscopic X-ray image will not match structurally across the entire image, and therefore, automatic positioning calculations may not yield sufficient accuracy.

[0008] Therefore, automatic positioning is performed on the fluoroscopic X-ray image, focusing only on the target structure. For this purpose, a medical professional sets the region where the target structure exists on the fluoroscopic X-ray image or on the pseudo-fluoroscopic X-ray image as the Region of Interest (ROI) beforehand. Then, during automatic positioning, the calculation is performed only on the image within the ROI, thereby achieving positioning that focuses only on the target structure within the ROI. The process of setting the ROI on the image is usually achieved by the medical professional operator drawing the ROI on the image.

[0009] Patent Document 1 describes a positioning device for patient positioning that creates a weighted region of interest using a fluoroscopic X-ray image taken on the day of treatment, based on a region of interest set in a pseudo-fluoroscopic X-ray image. Patent Document 2 describes a treatment system that compares the effective region of a volume image with a comparison image and performs positioning using the effective region (which can be replaced with the region of interest) updated by the comparison with the comparison image.

[0010] Japanese Patent Publication No. 2022-154279 Japanese Patent Publication No. 2021-061895

[0011] The technology disclosed in Patent Document 1 is effective in that it allows for precise alignment of specific areas of interest by determining the degree of importance for each specific part within the region of interest, which changes daily, and using the degree of importance as a weighting coefficient in the similarity calculation within the region of interest.

[0012] However, the area of ​​interest discussed here is a two-dimensional region set in advance during treatment planning, and does not take into account the amount of displacement the patient experiences on each treatment day.

[0013] The technology described in Patent Document 2 updates the effective region by aligning each small area within the effective region to the image taken on the day of treatment. Patent Document 2 does not take into account the amount of patient displacement that occurs each day of treatment, and requires new optimization calculations within the image to update the effective region.

[0014] If the patient's position deviates significantly from the planned location during fluoroscopic X-ray imaging on the day of treatment, it is possible that only a portion of the target structure will exist within the fluoroscopic X-ray image area corresponding to the ROI set on the pseudo-fluoroscopic X-ray image. In this case, the alignment calculation will start using only limited area information, leading to a deterioration in the accuracy of the final positioning calculation.

[0015] One way to avoid this problem is for the operator to use the manual alignment function beforehand to bring the simulated fluoroscopic X-ray image closer to the current fluoroscopic X-ray image before starting the automatic calculation. However, in this case, the operator's workload increases due to the manual alignment process.

[0016] Another workaround is to perform a rough positioning calculation using the entire image instead of the region of interest at the start of the calculation, and then perform the calculation using only the information within the region of interest. However, in this case, the computational load increases due to using the entire image, and the calculation accuracy is limited because areas that should not be included are included in the calculation domain.

[0017] This disclosure provides a bed position control device, a bed position control method, and a radiotherapy apparatus that enable more efficient positioning and control of the position of the patient bed.

[0018] To solve the above problems, the bed position control device according to this disclosure is a bed position control device that controls the position of a bed on which a patient is placed, and comprises: an image acquisition unit that acquires a fluoroscopic X-ray image of the patient taken by X-ray fluoroscopy; a pseudo-fluoroscopic X-ray image acquisition unit that acquires a pseudo-fluoroscopic X-ray image created from a three-dimensional image of the patient; a second region of interest creation unit that creates a second region of interest including the region of a target structure from a first region of interest set in the pseudo-fluoroscopic X-ray image to include the region of a target structure set to determine the position of the patient, in accordance with the patient's positional displacement; and an image matching unit that compares an image in the second region of interest from the pseudo-fluoroscopic X-ray image with an image in the second region of interest from the fluoroscopic X-ray image, calculates the position of the bed necessary to match the position of the patient in the fluoroscopic X-ray image, and moves the bed to the calculated position of the bed.

[0019] According to this disclosure, a second region of interest, including the area of ​​the target structure, is created in response to the patient's positional displacement. This allows the patient's position to be calculated without unnecessarily expanding the image matching range, and thus the position of the bed can be controlled.

[0020] Overall configuration diagram of the radiation therapy system. Flowchart of patient positioning process. Explanatory diagram showing a method for creating a second region of interest and an example of a GUI (Graphical User Interface). Explanatory diagram following Figure 3. Explanatory diagram schematically showing a method for setting a second region of interest. Explanatory diagram relating to Example 2, showing a method for setting a second region of interest using a three-dimensional region of interest and an example of a GUI. Explanatory diagram following Figure 6. Flowchart of patient positioning process relating to Example 3. Explanatory diagram relating to Example 4, showing a method for determining the patient's position based on a non-region of interest that is not a region of interest and an example of a GUI. Explanatory diagram following Figure 9. Flowchart of patient positioning process relating to Example 5. Flowchart of patient positioning process relating to Example 6.

[0021] Embodiments of the present invention will be described below with reference to the drawings. The bed position control device according to this embodiment sets a second region of interest considering the possible positional displacement of the patient that may occur between the time of treatment planning and the time of treatment, compares images within the second region of interest to calculate the positional displacement, and outputs a control signal based on the calculated amount of positional displacement. As a result, the bed position control device can efficiently calculate the patient's positional displacement by reducing the amount of calculation required for positional displacement while suppressing the inclusion of unnecessary images in the second region of interest.

[0022] This disclosure includes a positioning device as an example of a "bed position control device" that conforms to the following aspects. The following description is an example for understanding purposes and does not limit the scope of the present invention.

[0023] "A positioning device that follows a single viewpoint is a positioning device that controls the position of a bed on which a patient is placed. It creates a pseudo-fluoroscopic X-ray image by projecting a three-dimensional image used in creating a treatment plan for irradiating a patient with radiation in radiotherapy onto a predetermined surface. It obtains a first region of interest that includes the area of ​​the patient's target structure in the pseudo-fluoroscopic X-ray image. It sets a second region of interest by expanding the first region of interest, taking into account the three-dimensional movement of the target structure that may occur on the day of treatment. Based on the fluoroscopic X-ray image taken of the patient by X-ray fluoroscopy, the pseudo-fluoroscopic X-ray image, and the second region of interest, it calculates two-dimensional similarity information between the fluoroscopic X-ray image and the pseudo-fluoroscopic X-ray image. Based on the two-dimensional similarity information, it calculates the amount of movement required to move the bed so that the position of the target structure in the fluoroscopic X-ray image within the second region of interest coincides with the position of the target structure in the pseudo-fluoroscopic X-ray image." According to the above positioning device, highly accurate patient positioning can be easily achieved by setting a region of interest and positioning accordingly. Hereafter, the bed may be referred to as a couch.

[0024] Using Figures 1 to 5, we will explain the bed position control device according to Example 1 and the radiation therapy system as a "radiation therapy device" equipped with the bed position control device.

[0025] The following descriptions and drawings are illustrative examples for explaining the bed position control device according to this disclosure, and have been omitted and simplified as appropriate for clarity of explanation. This disclosure can also be implemented in various other forms. Unless otherwise specified, each component may be singular or plural. In the drawings illustrating embodiments, the same reference numeral is used for parts having the same function, and repeated explanations may be omitted. The position, size, shape, and range of each component shown in the drawings may not represent the actual position, size, shape, and range in order to facilitate understanding of the bed position control device of this disclosure. Therefore, the bed position control device and radiotherapy system according to this disclosure are not limited to the position, size, shape, and range disclosed in the drawings. When there are multiple identical or similar components, they may be described using the same reference numeral with different subscripts. However, if it is not necessary to distinguish between these multiple components, the subscript may be omitted.

[0026] The radiotherapy system of this embodiment is, for example, a particle beam therapy system. The radiotherapy system is a group of devices for irradiating a target patient with a particle beam. After positioning the patient, the radiotherapy system irradiates the patient with the particle beam.

[0027] In positioning the patient, the radiotherapy system aligns the patient with a digitally reconstructed radiography (DRR) image created by projecting three-dimensional patient image information using a second region of interest. The second region of interest is a region created by deforming a pre-drawn first region of interest, taking into account the amount of patient displacement that may occur during treatment. The patient bed position control device 20 used in the radiotherapy system 1000 includes, for example, a digitally reconstructed radiography image creation unit 22, a region of interest drawing unit 23, a second region of interest creation unit 24, an image matching unit 25, and an image display unit 26. The patient bed position control device 20 may also include functional units other than those 22, 23, 24, 25, and 26, as described later. A functional unit in this disclosure is hardware, software, or a device in which hardware and software cooperate to realize a predetermined function.

[0028] The pseudo-fluoroscopic X-ray image creation unit 22, which functions as a "pseudo-fluoroscopic X-ray image acquisition unit," creates a pseudo-fluoroscopic X-ray image by projecting a three-dimensional image used to create a treatment plan for irradiating a patient with radiation in radiotherapy onto a predetermined surface.

[0029] The region of interest drawing unit 23 sets the region to be aligned on the pseudo-fluoroscopic X-ray image as the first region of interest.

[0030] The second region of interest creation unit 24 creates a second region of interest, which includes the area of ​​the target structure set to determine the patient's position, from the first region of interest drawn on the pseudo-fluoroscopic X-ray image to correspond to the patient's positional displacement.

[0031] The image matching unit 25 calculates the similarity between the fluoroscopic X-ray image and the image within the second region of interest created from the pseudo-fluoroscopic X-ray image. The image matching unit 25 calculates the amount of movement required to move the examination table so that the position of the target structure in the fluoroscopic X-ray image matches the position of the target structure in the pseudo-fluoroscopic X-ray image.

[0032] The image display unit 26, acting as a "display unit," provides the user (a medical professional acting as the operator) with information including the matching results from the image matching unit 25.

[0033] Figure 1 shows the overall configuration of the radiotherapy system 1000 according to Embodiment 1. The radiotherapy system 1000 of this embodiment is a system that treats patients using particle beams and comprises, for example, an accelerator 1, a beam transport device 2, a gantry 3, an irradiation nozzle 4, planar detectors 5A, 5B, X-ray tubes 6A, 6B, a patient bed 7, a robotic arm 8, a communication device 9, a data server 10, a treatment planning device 11, a fluoroscopic X-ray image acquisition device 12, a patient bed control device 13, and a patient positioning device 20.

[0034] The patient positioning device 20, which functions as a "bed position control device" (hereinafter sometimes referred to as the "positioning device"), includes, for example, an image acquisition unit 21, a pseudo-fluoroscopic X-ray image creation unit 22, a region of interest drawing unit 23, a second region of interest creation unit 24, an image matching unit 25, an image display unit 26, and a control unit 27.

[0035] Before particle beam therapy, the following steps are performed to position the patient. First, the patient PA is placed and secured on the treatment table 7. Next, a robotic arm 8 connected to the treatment table 7 moves the treatment table 7 to a pre-planned position. The planned position refers to the position of the patient PA in the radiation therapy room that replicates the patient placement in the pre-created treatment plan. The robotic arm 8 can be driven in a total of six axes, for example, in three translational directions and three rotational directions. The robotic arm 8 can position the treatment table 7 with the patient PA at the appropriate position and angle. These steps constitute patient positioning for particle beam therapy.

[0036] The particle beam used for treatment is generated in accelerator 1. The particle beam is accelerated in accelerator 1 to an energy suitable for treatment, and then transported to gantry 3 by beam transport device 2. Gantry 3 has a rotation mechanism. Gantry 3 can rotate 360° so that the particle beam can be irradiated to the affected area of ​​the patient PA at any angle. Gantry 3 deflects the particle beam in the appropriate direction and irradiates the main body of the patient PA from irradiation nozzle 4.

[0037] The irradiation nozzle 4 incorporates a mechanism (not shown) that changes the shape of the particle beam to match the shape of the affected area of ​​the patient's PA.

[0038] The treatment planning device 11 creates a treatment plan based on three-dimensional image information (CT images) of the patient's PA before irradiating the patient with particle beams. The treatment planning device 11 acquires the three-dimensional image information of the patient's PA from the data server 10 via the communication device 9. The treatment planning device 11 calculates the appropriate irradiation angle, irradiation shape, and irradiation dose of the particle beam from the acquired three-dimensional image, and determines these calculated parameters as irradiation information. The treatment planning device 11 saves the determined irradiation information to the data server 10 via the communication device 9.

[0039] The fluoroscopic X-ray imaging device 12 acquires fluoroscopic X-ray images by controlling the flat-panel detector 5A and X-ray tube 6A, and the flat-panel detector 5B and X-ray tube 6B, which are arranged in mutually orthogonal directions. The acquired fluoroscopic X-ray images are sent to the patient positioning device 20.

[0040] The patient positioning device 20 as a "sleeper position control device" calculates the movement amount for moving the sleeper 7 so that the position of the patient PA when the treatment plan is created coincides with the position of the patient PA arranged on the sleeper 7 for irradiating the particle beam, from the fluoroscopic X-ray image and the pseudo-fluoroscopic X-ray image.

[0041] The patient positioning device 20 acquires the irradiation information created at the time of treatment planning and the three-dimensional image information of the patient PA from the data server 10 via the communication device 9.

[0042] The image acquisition unit 21 of the patient positioning device 20 controls the fluoroscopic X-ray image imaging device 12 to acquire a fluoroscopic X-ray image.

[0043] The pseudo-fluoroscopic X-ray image creation unit 22 of the patient positioning device 20 creates a pseudo-fluoroscopic X-ray image by arranging and projecting the three-dimensional image of the patient PA on the same virtual space as the actual fluoroscopic X-ray imaging system.

[0044] The three-dimensional image information includes information indicating the shape and electron density of the patient in voxel units. Generally, a computed tomography (CT) imaging image is used as the three-dimensional image information.

[0045] The region of interest drawing unit 23 causes an operator to draw, on the screen, a region including a target structure (for example, bone) to be used for alignment on the pseudo-fluoroscopic X-ray image as a first region of interest.

[0046] The second region of interest creation unit 24 creates, corresponding to the patient's positional deviation, a second region of interest including the region of the target structure from the first region of interest set in the pseudo-fluoroscopic X-ray image.

[0047] The image matching unit 25 matches the image in the second region of interest in the fluoroscopic X-ray image with the image in the second region of interest in the pseudo-fluoroscopic X-ray image, and calculates the displacement amount (positional deviation amount) of six degrees of freedom necessary to match the positions of the patient in both images. The calculated displacement amount is transmitted as a control signal from the positioning device 20 to the sleeper control device 13. The sleeper control device 13 moves the sleeper 7 to a predetermined position according to the control signal received from the positioning device 20.

[0048] The image display unit 26, for example, displays the matched image on a monitor display (not shown) by associating it with the first region of interest and the second region of interest. The image output destination is not limited to the monitor display; it may also be a head-mounted display or a printer.

[0049] The image matching described above is performed based on the similarity between the fluoroscopic X-ray image and the pseudo-fluoroscopic X-ray image within the second region of interest. As an example of similarity information, the zero-mean normalized cross-correlation (ZNCC) coefficient is explained. The ZNCC similarity can be calculated using Equation 1. In Equation 1, g(i,j) is the DR image, f(i,j) is the DRR image, μg is the average brightness value of the DR image, and μf is the average brightness value of the DRR image.

[0050] The patient positioning device 20 consists of various information processing devices, such as a computer. The information processing device includes, for example, a arithmetic unit such as a processor, a storage device such as memory, and a communication interface (not shown). The information processing device may also include an information input / output device (not shown) for exchanging information with the user.

[0051] The arithmetic unit is, for example, a CPU (Central Processing Unit), an FPGA (Field-Programmable Gate Array), etc. The arithmetic unit may also include dedicated circuits that perform only specific calculations.

[0052] Storage devices include, for example, magnetic storage media such as HDDs (Hard Disk Drives), and semiconductor storage media such as RAM (Random Access Memory), ROM (Read Only Memory), and SSDs (Solid State Drives). Furthermore, storage devices may include devices that can attach and detach storage media such as optical discs, magneto-optical discs, magnetic discs, and magnetic tapes. Data other than computer programs may be stored in the storage media. The storage media stores computer programs and / or data non-temporarily and is readable and writable by an information processing device (computer).

[0053] The storage device stores, for example, computer programs such as firmware and an operating system, and predetermined computer programs for realizing the functions of the patient positioning device 20.

[0054] The computing unit reads a computer program from its storage device and executes it when the positioning device 20 starts operating (for example, when the power is turned on), thereby realizing the functions of each functional unit 21 to 27 of the positioning device 20. This enables the execution of a series of controls as a patient positioning device 20.

[0055] Information input / output devices that exchange information with users include information input devices and information output devices. Information input devices include, for example, pointing devices such as mice, keyboards, touch panels, eye-tracking devices, and voice input devices. Information output devices include, for example, monitor displays, head-mounted displays, and printers. An information input / output device may also be a touch panel that performs both information input and output.

[0056] The patient positioning device 20 may be a so-called on-premise device installed in or near the treatment room, or it may be a cloud computer located on a communication network.

[0057] The radiotherapy system 1000 creates a first region of interest before positioning the patient for particle beam therapy. Furthermore, the radiotherapy system 1000 deforms the first region of interest based on a pre-set amount of patient displacement to create a second region of interest, and calculates the amount of displacement by comparing images within the second region of interest. The method for creating the region of interest and the method for positioning the patient in this embodiment will be described below with reference to Figures 2 to 7.

[0058] Figure 2 is a flowchart illustrating the patient positioning process, which uses a second region of interest to determine the patient's position. The patient positioning process calculates the difference (positional deviation) between the patient's current position and their ideal position, and then calculates a control quantity to eliminate that difference.

[0059] In the figures, the pseudo-fluoroscopic X-ray image is displayed as the DRR image, the fluoroscopic X-ray image as the DR image, the first region of interest as the first ROI, and the second region of interest as the second ROI. Figures 3 and 4 show the method for creating the second region of interest.

[0060] As shown in Figure 2, the pseudo-fluoroscopic X-ray image creation unit 22 of the positioning device 20 creates a pseudo-fluoroscopic X-ray image by retrieving the patient's planned CT image 200 stored in either the treatment planning device 11 or the data server 10 and performing projection calculations (S100).

[0061] As shown in Figure 3, the pseudo-fluoroscopic X-ray image creation unit 22 arranges virtual X-ray sources 202A, 202B and virtual plane detectors 201A, 201B in the virtual space in the same layout as the actual X-ray sources and plane detectors. The units are then arranged so that the reference position of the planned CT image 200 (for example, the isocenter position) coincides with the intersection point P1 where the projection centerlines connecting the centers of each virtual X-ray source 202A, 202B and the virtual plane detectors 201A, 201B intersect.

[0062] As shown in Figure 4, the pseudo-fluoroscopic X-ray image creation unit 22 calculates the intensity of virtual X-rays incident on each pixel of the virtual flat-panel detectors 201A and 201B, taking into account the attenuation of X-rays when virtual X-rays emitted from virtual X-ray sources 202A and 202B are incident on the virtual flat-panel detectors 201A and 201B through projection processing. As a result, the pseudo-fluoroscopic X-ray image creation unit 22 creates pseudo-fluoroscopic X-ray images 206A and 206B.

[0063] Next, the region of interest drawing unit 23 sets the regions on the pseudo-fluoroscopic X-ray images 206A and 206B in which the target structure to be used for positioning exists as the first regions of interest 203A and 203B, based on the user's operation (S101).

[0064] An example of how to set the first regions of interest 203A and 203B will be explained. One method is for the region of interest drawing unit 23 to display pseudo-fluoroscopic X-ray images 206A and 206B on the screen and have the user fill in the areas to be designated as the first regions of interest 203A and 203B using mouse operations or the like. The region of interest drawing unit 23 can then set the areas specified by the user (the filled areas) as the first regions of interest 203A and 203B.

[0065] Alternatively, the region of interest drawing unit 23 can display a simulated fluoroscopic X-ray image on the screen and allow the user to select a position within the region to be designated as the first region of interest 203A, 203B using mouse operations or the like. The region of interest drawing unit 23 can also set the region within a predetermined distance from the position selected by the user as the first region of interest 203A, 203B.

[0066] As yet another method, the region of interest drawing unit 23 displays pseudo-fluoroscopic X-ray images 206A and 206B on the screen and allows the user to select a position within the region to be designated as the first region of interest 203A and 203B using mouse operations or the like. The region of interest drawing unit 23 can set a continuous region having pixel values ​​within a certain range from the pixel value of the position selected by the user as the first region of interest 203A and 203B. Any of the methods described above may be used. Alternatively, the first region of interest may be set on the pseudo-fluoroscopic X-ray image using a method other than those described above.

[0067] Return to Figure 2. The user sets the amount of displacement (deformation) of the patient (more precisely, the target structure such as the patient's bones) from the planned position that may occur on the treatment day (S102). The amount of displacement may be set manually by the user. Alternatively, a fixed or variable value calculated from the type and position of the target structure may be used as the amount of displacement.

[0068] The image matching unit 25 of the positioning device 20 reads the first regions of interest 203A and 203B created by the region of interest drawing unit 23 (S103). Based on the set range of movement amounts, the image matching unit 25 determines the translational and rotational movement amounts of the first regions of interest drawn on the pseudo-fluoroscopic X-ray image.

[0069] The image matching unit 25 calculates the translation amount generated by the amount of movement on the pseudo-fluoroscopic X-ray image created by projecting onto a virtual plane detector, for example, as follows.

[0070] Let d = (x, y, z)^T be the amount of movement set in the treatment room space. If the imaging angle in the gantry rotation direction is φ and the rotation angle in the Yaw direction is θ in the imaging device in the treatment room space, then the rotation matrices Ry and Rz in each direction are given by Equation 2. Let ex and ey be the unit direction vectors in the x and y directions (the "x" and "y" directions shown on the virtual plane detector 201B in Figure 3) of the virtual plane detector (pseudo-fluoroscopic X-ray image) in the treatment room space when the imaging angles φ = 0° and θ = 0°. The unit direction vectors ex and ey are given by Equation 3.

[0071] Note that ey includes a negative directional component because in a two-dimensional planar coordinate system, the direction opposite to the y direction in the treatment room space is considered positive. If the unit direction vectors when the rotation angles of the imaging direction are φ and θ are ex' and ey' (see Equation 3), then the translational displacement in the x and y directions of the region of interest on the pseudo-fluoroscopic X-ray image created by projection processing onto the virtual plane detector 201 is (x ｓｈｉｆｔ , y ｓｈｉｆｔ ) can be found using mathematical formula 4.

[0072]

[0073]

[0074] The image matching unit 25 determines the amount of movement on the pseudo-fluoroscopic X-ray image, and then deforms the first region of interest on the pseudo-fluoroscopic X-ray image by the calculated translation and rotation amounts. The image matching unit 25 determines the regions when translated and rotated for the range of the movement amount, and sets the regions that include all of the translated and rotated regions as the second regions of interest 205A and 205B, as shown in Figure 4 (S104).

[0075] As an example, the image matching unit 25 sets the sum of the regions obtained by translating and rotating the first region of interest by the amount of movement as the second region of interest. However, as will be described later, the image matching unit 25 may also set the region obtained by translating the first region of interest by the amount of movement as the second region of interest, or it may set the region obtained by rotating the first region of interest by the amount of movement as the second region of interest.

[0076] Returning to Figure 2, the fluoroscopic X-ray imaging device 12 acquires a fluoroscopic X-ray image (DR image) of the patient PA (S105). The image display unit 26 displays the first regions of interest 203A, 203B and the second regions of interest 205A, 205B, created by the region of interest drawing unit 23, on top of the pseudo-fluoroscopic X-ray images 206A, 206B and the fluoroscopic X-ray images 207A, 207 (see Figure 4).

[0077] The fluoroscopic X-ray image acquired by the fluoroscopic X-ray imaging device 12 is transmitted to the patient positioning device 20. The image matching unit 25 automatically performs calculations to match the position of the image within the second region of interest with the pseudo-fluoroscopic X-ray image (S106).

[0078] In this embodiment, as an example, a normalized cross-correlation coefficient or cross-information, which are commonly used as indicators for measuring similarity, is used. The image matching unit 25 optimizes the displacement amounts (three translational directions and three rotational directions) of the six degrees of freedom of the pseudo-fluoroscopic X-ray image or the fluoroscopic X-ray image, which are necessary for the pseudo-fluoroscopic X-ray image and the image of the second region of interest to match. Indicators other than those mentioned above may also be used to measure similarity.

[0079] Furthermore, in step S106, the image matching unit 25 determines whether the calculated similarity satisfies the pre-set convergence conditions. If the calculated similarity does not satisfy the convergence conditions, the image matching unit 25 searches for conditions that satisfy the convergence conditions (values ​​for a total of 6 degrees of freedom: 3 translational degrees of freedom and 3 rotational degrees of freedom) through optimization processing. Examples of multivariable optimization processing methods such as 6 degrees of freedom include the sloping simplex method or the Powell method. The image matching unit 25 can use these methods as appropriate. The image matching unit 25 may also perform optimization processing using other methods.

[0080] Through the above steps, the image matching unit 25 calculates the amount of movement required to position the patient in the planned location. The calculated amount of movement is sent to the control unit 27. The control unit 27 outputs a control signal to the bed control device 13, which moves the couch by the amount of movement (S107). This completes the patient positioning process.

[0081] Figure 5 is a schematic diagram illustrating how to set up the second region of interest. At the top of Figure 5, a method for creating the second region of interest 205(1) by translating the first region of interest 203(1) is shown. When the first region of interest 203(1) is moved by an amount MH in the horizontal direction in the figure, the first region of interest 204(1) is obtained.

[0082] The second region of interest 205(1) is set to include both the (original) first region of interest 203(1) before the move and the first region of interest 204(1) after the move. The second region of interest 205(1) can be set outside the region formed by combining the first region of interest 203(1) before the move and the first region of interest 204(1) after the move, with a small margin M1.

[0083] The central part of Figure 5 shows a method for creating the second region of interest 205(2) by rotating the first region of interest 203(2). When the first region of interest 203(2) is rotated clockwise by an amount MA (rotation angle MA) in the figure, the first region of interest 204(2) is obtained.

[0084] The second region of interest 205(2) is set to include both the (original) first region of interest 203(2) before the move and the first region of interest 204(2) after the move. The second region of interest 205(2) can be set outside the region formed by combining the first region of interest 203(2) before the move and the first region of interest 204(2) after the move, with a small margin M1.

[0085] The bottom of Figure 5 shows a method for creating the second region of interest 205(3) by translating and rotating the first region of interest 203(3). When the first region of interest 203(3) is moved horizontally by an amount MH in the figure, the first region of interest 204(3-1) is obtained. When the first region of interest 203(3) is rotated clockwise by an amount MA in the figure, the first region of interest 204(3-2) is obtained.

[0086] The second region of interest 205(3) is set to include all three regions: the first region of interest 203(3) before movement, the first region of interest 204(3-1) after translational movement, and the first region of interest 204(3-2) after rotation. The second region of interest 205(3) can be set outside the region formed by combining the first region of interest 203(1) before movement, the second region of interest 204(3-1) after translation, and the first region of interest 204(3-2) after rotation, with a small margin M1.

[0087] The margin M1 may be a fixed value or a variable value. The margin M1 can be set appropriately according to the shapes of the first regions of interest 203 and 204 before movement. The shape of the second region of interest 205 can be any shape, such as a roughly rectangular shape, a roughly elliptical shape, or a polygonal shape.

[0088] According to this embodiment, even when the position of the target structure on the patient is far from the planned position, the patient's position can be calculated using only the region under the second region of interest that includes both the target structure on the pseudo-fluoroscopic X-ray image and the target structure on the fluoroscopic X-ray image. Therefore, the patient positioning device 20 of this embodiment can determine the patient's position with high accuracy without increasing the computational load.

[0089] Example 2 will be described using Figures 6 and 7. In the following examples, including this one, the differences from Example 1 will be described in particular.

[0090] In Example 1, the second region of interest was set based on the region included when the first region of interest was translated and rotated in two dimensions. In this example, the patient positioning device 20 uses the first region of interest to calculate the region of interest in three-dimensional space (hereinafter referred to as the three-dimensional region of interest). The patient positioning device 20 sets the two-dimensional region that is projected after translating and rotating the three-dimensional region of interest based on the set range of movement as the first region of interest. In addition to cases where the first region of interest is translated and rotated, cases where only the first region of interest is translated, or cases where only the second region of interest is rotated, are also included in this example, as in Example 1.

[0091] Using Figures 6 and 7, a method for creating second regions of interest 205A and 205B using the three-dimensional region of interest 301 will be explained. Similar to Figure 3, the patient positioning device 20 places the planned CT image 200 in the virtual space and creates pseudo-fluoroscopic X-ray images 206A and 206B through projection processing. The user manually sets the first regions of interest 203A and 203B on the pseudo-fluoroscopic X-ray images 206A and 206B.

[0092] The patient positioning device 20 calculates cone regions (hereinafter referred to as region of interest cones 300A, 300B) that connect the contours of the first region of interest 203A, 203B drawn on the virtual plane detectors 201A, 201B (pseudo-fluoroscopic X-ray images 206A, 206B) in the virtual space with the virtual X-ray sources 202A, 202B.

[0093] The patient positioning device 20 sets the region where the two cones of interest 300A and 300B overlap as the three-dimensional region of interest 301. The patient positioning device 20 translates and rotates the three-dimensional region of interest 301 based on a pre-set range of movement. The patient positioning device 20 sets the region of interest 302 formed on the virtual plane detectors 201A and 201B (pseudo-fluoroscopic X-ray images 206A and 206B) when the three-dimensional region of interest 301 is translated and rotated as the second region of interest.

[0094] The second region of interest established through the above process can also be used for positioning between the fluoroscopic X-ray image and the pseudo-fluoroscopic X-ray image.

[0095] This embodiment provides the same effects as in Embodiment 1. Furthermore, in this embodiment, the region of interest can be set considering the actual movement of the target structure, rather than when the first region of interest, which is set in advance by the user, is translated and rotated on a two-dimensional image.

[0096] Example 3 will be explained using Figure 8. In Example 1, it is assumed that the patient positioning calculation result is correct (S106 in Figure 2). However, in this example, a precaution is taken in the case where the patient positioning device 20 is unable to perform the initial positioning calculation with an accuracy of a predetermined value or higher.

[0097] Figure 8 is a flowchart of the patient positioning process according to this embodiment. Steps S100 to S107 shown in Figure 8 are the same as steps S100 to S107 described in Figure 2, so their explanation is omitted. In this embodiment, new steps S110 and S111 are added between steps S106 and S107.

[0098] In this embodiment, the patient positioning device 20 determines, after the automatic calculation in step S106, whether the positioning accuracy calculated in step S106 is equal to or greater than a predetermined value (S110). Whether the calculation result (positioning value) in step S106 is equal to or greater than a predetermined accuracy may be automatically performed by, for example, the image matching unit 25 of the patient positioning device 20, or a user may make a visual judgment and input the judgment result to the patient positioning device 20. The image matching unit 25 can also determine that the accuracy of the positioning calculation is less than a predetermined value from the similarity value of the optimization process.

[0099] If the image matching unit 25 determines that the accuracy of the positioning calculation is less than a predetermined value (S110: NO), it adjusts the second region of interest (S111), and then repeats the process in step S106 using the adjusted second region of interest.

[0100] There are several possible methods for making adjustments in step S111. The first method is to manually edit the user-drawn region of interest to create a second region of interest. The second method is to reset the amount of deformation that may occur in a clinical setting to create a second region of interest. The third method is to perform both the first and second methods to create a second region of interest.

[0101] A fourth method involves directly editing the second domain of interest. A fifth method involves using artificial intelligence. In this method, for example, the AI ​​can convert the user's utterances into commands based on past knowledge and automatically set a new second domain of interest.

[0102] Embodiment 4 will be explained using Figures 9 and 10. In Embodiment 1, the user sets a region of interest on the simulated fluoroscopic X-ray image as the first region of interest, the second region of interest creation unit 24 creates the second region of interest based on the amount of movement information that may occur on the treatment day, and the image matching unit 25 calculates the similarity using only the images within the second region of interest and matches the images.

[0103] In contrast, in this embodiment, areas that are not of interest are drawn on the pseudo-fluoroscopic X-ray image and designated as the "first region of non-interest." The patient positioning device 20 in this embodiment creates a "second region of non-interest" from the first region of non-interest and the amount of movement that may occur on the treatment day. The image matching unit 25 in this embodiment uses images within areas other than the second region of non-interest to calculate the similarity.

[0104] The method for creating the second region of disinterest and the positioning calculation using the second region of disinterest will be explained below using Figures 9 and 10.

[0105] As shown in the pseudo-fluoroscopic X-ray images 206A and 206B in Figure 10, all but the three in the middle of the seven vertebrae are expected to deform during treatment. The patient positioning device 20 sets first regions of no interest 400A and 400B for the four bones where deformation may occur. Next, the user sets the amount of displacement of the patient (target structure) from the planned position that may occur on the day of treatment.

[0106] The patient positioning device 20 determines the amount of movement on the pseudo-fluoroscopic X-ray image, and then determines the regions 401A and 401B obtained by deforming the first region of inconcern by that amount of movement (translational and rotational), for the entire range of movement, and sets the region containing all of these as the second region of inconcern 402A and 402B.

[0107] Another method for creating the second regions of disinterest 402A and 402B is to translate and rotate the three-dimensional region of disinterest, and then project the three-dimensional region of disinterest onto the virtual plane detectors 201A and 201B (pseudo-fluoroscopic X-ray images 206A and 206B) to set the region of disinterest as the second region of disinterest.

[0108] The image display unit 26 can also overlay the first non-interesting regions 400A, 400B and the second non-interesting regions 402A, 402B, which are set by the user, onto the pseudo-fluoroscopic X-ray images 206A, 206B and the acquired fluoroscopic X-ray images 207A, 207. Subsequently, the acquired fluoroscopic X-ray images are transmitted to the patient positioning device 20. The image matching unit 25 matches the positions of the pseudo-fluoroscopic X-ray images and images other than the second non-interesting regions.

[0109] Example 5 will be explained using Figure 11. In this example, the patient positioning device 20 creates a second region of interest when creating a treatment plan.

[0110] The creation of the second region of interest and its application during patient positioning in this embodiment will be explained using Figure 11. The contents of steps S300 to S303 are the same as steps S100 to S103 explained in Figure 2 of Embodiment 1. The patient positioning device 20 of this embodiment performs steps S300 to S304, enclosed by the dotted line in Figure 11, either during treatment planning or during the period from the creation of the treatment plan to the treatment day.

[0111] The pseudo-fluoroscopic X-ray image creation unit 22 of the positioning device 20 creates a pseudo-fluoroscopic X-ray image by retrieving the patient's planned CT image stored in the treatment planning device 11 or data server 10 and performing projection calculations (S300).

[0112] Based on user input, the region of interest drawing unit 23 sets the regions on the pseudo-fluoroscopic X-ray image where the target structure to be used for positioning is located as the first regions of interest 203A and 203B (S301).

[0113] The user sets the amount of displacement of the patient (target structure) from the planned position that may occur on the treatment day (S302).

[0114] The image matching unit 25 reads the first regions of interest 203A and 203B created by the region of interest drawing unit 23 (S303). Then, the image matching unit 25 determines the translation and rotation amounts of the first regions of interest drawn on the pseudo-fluoroscopic X-ray image based on the set range of movement amounts.

[0115] As described in Example 1, the image matching unit 25, after determining the amount of movement on the image, determines the area where the first region of interest is deformed on the image by the calculated translational and rotational amounts, for the range of the movement amount, and sets the area containing all of these as the second region of interest 205A, 205B (S304). Furthermore, the image matching unit 25 causes the second region of interest 205A, 205B to be saved as region of interest information in the storage area of ​​the positioning device 20, the data server 10, or the treatment planning device 11 (S304).

[0116] When positioning the patient on the treatment day, the positioning device 20 reads the stored region of interest information (S305). The fluoroscopic X-ray imaging device 12 acquires a fluoroscopic X-ray image of the patient PA (S306). The image display unit 26 overlays the first regions of interest 203A, 203B and the second regions of interest 205A, 205B, which were created by the user drawing in the region of interest drawing unit 23, onto the pseudo-fluoroscopic X-ray images 206A, 206B and the acquired fluoroscopic X-ray images 207A, 2067.

[0117] The acquired fluoroscopic X-ray image is transmitted to the positioning device 20. The image matching unit 25 performs automatic calculations for position matching between the pseudo-fluoroscopic X-ray image and the image in the second region of interest (S307). For example, using a normalized cross-correlation coefficient or mutual information, the unit optimizes and calculates the displacement amounts (three translational directions and three rotational directions) of the six degrees of freedom of the pseudo-fluoroscopic X-ray image or fluoroscopic X-ray image necessary for the two images to match. Other indicators may be used for similarity.

[0118] The image matching unit 25 determines whether the calculated similarity satisfies the pre-set convergence conditions. If the similarity does not satisfy the convergence conditions, the image matching unit 25 searches for conditions that satisfy the convergence conditions (for example, values ​​for a total of 6 degrees of freedom, with 3 translational degrees of freedom and 3 rotational degrees of freedom) through optimization processing. As a method for multivariable optimization processing such as 6 degrees of freedom, the image matching unit 25 can employ methods such as the sloping simplex method or the Powell method. However, optimization processing may be performed using methods other than these.

[0119] Through the above steps, the amount of couch movement required to position the patient in the planned location is calculated. The control unit 27 outputs a control signal to the bed control device 13, which moves the couch by the amount of couch movement (S308). This completes the patient positioning process.

[0120] There are advantages and disadvantages to creating the second area of ​​interest during the treatment planning stage, or immediately before the start of the first treatment (immediately after fluoroscopic X-ray imaging).

[0121] When drawing the second region of interest on a pseudo-fluoroscopic X-ray image (DRR) and setting the displacement (movement amount) based on the fluoroscopic X-ray image (DR) taken on the first day of a series of treatments for a patient, the advantage is that the displacement amount can be set to suit the patient, thus improving positioning accuracy. On the other hand, the disadvantage is that the workload on the treatment day increases.

[0122] Setting the second region of interest and displacement amount based on the user's past knowledge during treatment planning has the advantage of reducing the workload on treatment days, as there is no need to set the second region of interest on the treatment day. On the other hand, a disadvantage is that it may be necessary to readjust the second region of interest and displacement amount depending on the patient's condition on the treatment day.

[0123] Example 6 will be explained using Figure 12. In this example, the patient positioning device 20 utilizes AI (artificial intelligence) to understand user instructions and to match images.

[0124] The patient positioning process in Figure 12 is similar to the patient positioning process described in Figure 2, but it includes several steps involving AI.

[0125] The pseudo-fluoroscopic X-ray image creation unit 22 creates a pseudo-fluoroscopic X-ray image by retrieving the patient's planned CT image 200 stored in either the treatment planning device 11 or the data server 10 and performing projection calculations (S400).

[0126] The region of interest drawing unit 23 sets the region where the target structure to be used for positioning is located on the pseudo-fluoroscopic X-ray image as the first region of interest using a machine learning model (S401). The machine learning model is generated by pre-learning information on pairs of a large number of pseudo-fluoroscopic X-ray images and the first regions of interest set in those pseudo-fluoroscopic X-ray images. The machine learning model can propose an appropriate first region of interest to the user from the input pseudo-fluoroscopic X-ray image. The positioning device 20 stores the first region of interest that has been approved by the user.

[0127] The positioning device 20 uses AI to set the amount of displacement of the patient (target structure) from the planned position that may occur on the treatment day (S402). The machine learning model for estimating the displacement is generated by learning from actual measurement data (images) of the displacement of target structures for a large number of patients. The positioning device 20 can set the amount of displacement of the target structure using this machine learning model and the user's knowledge.

[0128] The image matching unit 25 reads the pseudo-fluoroscopic X-ray image and the first region of interest created by the region of interest drawing unit 23 (S403).

[0129] The image matching unit 25 uses AI to deform the first region of interest on the pseudo-fluoroscopic X-ray image by the calculated translation and rotation amounts, and sets it as the second region of interest (S404).

[0130] The fluoroscopic X-ray imaging device 12 acquires a fluoroscopic X-ray image (DR image) of the patient PA (S405).

[0131] The image matching unit 25 automatically performs calculations to match the position of the image within the second region of interest with the pseudo-fluoroscopic X-ray image (S406). This calculation can also be performed by AI.

[0132] The user uses so-called generative AI to give instructions to the positioning device 20, which then outputs control signals to the bed control device 13 to adjust the position of the bed 7 (S407). For example, if the user inputs sentences such as "A little higher," "You've gone too far. A little lower," or "Tilt to the right" into the generative AI via a microphone, which is part of the information input / output device, the generative AI understands what the user has said and converts it into control signals.

[0133] This embodiment, configured in this way, also provides the same effects as the first embodiment. In this embodiment, since so-called AI is used to perform at least a part of the positioning process, the positioning process can be performed efficiently. It should be noted that it is not necessarily required to use AI in all of the steps S401-S407 shown in Figure 12. AI may be used in at least one of the steps S401-S407.

[0134] In this embodiment, a particle beam therapy system is given as an example of a radiation therapy system. However, as an example that is not limited to particle beam therapy, the accelerator 1 described in this embodiment may be an electron beam accelerator intended for therapeutic X-rays, or a particle beam accelerator such as a proton or carbon accelerator.

[0135] It is clear that this disclosure describes the following configurations to a degree that would be implementable by a person skilled in the art.

[0136] (Note 1) A bed position control device for controlling the position of a bed on which a patient is placed, comprising: an image acquisition unit that acquires a fluoroscopic X-ray image of the patient taken by X-ray fluoroscopy; a pseudo-fluoroscopic X-ray image acquisition unit that acquires a pseudo-fluoroscopic X-ray image created from a three-dimensional image of the patient; a second region of interest creation unit that creates a second region of interest including the region of a target structure set in the pseudo-fluoroscopic X-ray image to correspond to the positional displacement of the patient, from a first region of interest set in the pseudo-fluoroscopic X-ray image to include the region of a target structure set to determine the position of the patient; and an image matching unit that compares an image in the pseudo-fluoroscopic X-ray image within the second region of interest with an image in the fluoroscopic X-ray image within the second region of interest, calculates the position of the bed necessary to match the position of the patient in the fluoroscopic X-ray image, and moves the bed to the calculated position of the bed.

[0137] (Note 2) The bed position control device according to Note 1, wherein the second region of interest creation unit creates the second region of interest that encloses the first region of interest by deforming the first region of interest.

[0138] (Note 3) The bed position control device according to Note 1 or 2, wherein the second region of interest is created based on the extent of the region obtained by translating the first region of interest.

[0139] (Note 4) The bed position control device according to any one of Notes 1 to 3, wherein the second region of interest is created based on the extent of the region obtained by rotating the first region of interest.

[0140] (Note 5) The bed position control device according to any one of Notes 1 to 4, wherein the second region of interest is created based on both the expansion of the region obtained by translating the first region of interest and the expansion of the region obtained by rotating the first region of interest.

[0141] (Note 6) The bed position control device according to any one of Notes 1 to 5, wherein the image matching unit causes the second region of interest creation unit to recreate the second region of interest when the calculation accuracy of the bed position falls below a predetermined accuracy.

[0142] (Note 7) The bed position control device according to any one of Notes 1 to 6, wherein the second region of interest creation unit assumes the fluoroscopic X-ray image, defines the region formed by the overlap of the first region of interest and the X-ray source in three-dimensional space as the three-dimensional region of interest, deforms the three-dimensional region of interest in accordance with the patient's positional displacement, and creates the region formed by projecting the deformed three-dimensional region of interest onto a plane as the second region of interest.

[0143] (Note 8) The second region of interest creation unit is a bed position control device according to any one of Notes 1 to 7 that creates the second region of interest on the day the patient is to receive radiation therapy.

[0144] (Note 9) The bed position control device described in any one of Notes 1 to 8, wherein the second region of interest creation unit creates and saves the second region of interest when creating a treatment plan for radiation therapy to the patient.

[0145] (Note 10) A bed position control device according to any one of Notes 1 to 9, further comprising a display unit that outputs a relationship between the first region of interest, the second region of interest, and the relationship between the position of the target structure in the fluoroscopic X-ray image and the position of the target structure in the pseudo-fluoroscopic X-ray image.

[0146] (Note 11) A method for controlling the position of a bed on which a patient is placed using a bed position control device, comprising: acquiring a fluoroscopic X-ray image of the patient taken by X-ray fluoroscopy; acquiring a pseudo-fluoroscopic X-ray image created from a three-dimensional image of the patient; creating a second region of interest including the region of the target structure from a first region of interest set in the pseudo-fluoroscopic X-ray image to include the region of the target structure, corresponding to the positional displacement of the patient; comparing the image in the second region of interest from the pseudo-fluoroscopic X-ray image with the image in the second region of interest from the fluoroscopic X-ray image; calculating the position of the bed necessary to match the position of the patient in the fluoroscopic X-ray image; and moving the bed to the calculated position of the bed.

[0147] (Note 12) A radiotherapy apparatus comprising a bed position control device described in any one of Notes 1 to 10, and a radiation irradiation device that irradiates a patient on a bed whose position is controlled by the bed position control device with predetermined radiation.

[0148] 1000...Radiation therapy system, 1...Accelerator, 2...Beam transport device, 3...Gantry, 4...Irradiation nozzle, 5A...Planar detector, 5B...Planar detector, 6A...X-ray tube, 6B...X-ray tube, 7...Patient table, 8...Robot arm, 9...Patient, 10...Treatment planning device, 11...Communication device, 12...Data server, 14...Fluoroscopy X-ray image acquisition device, 15...Patient table control device, 20...Patient positioning device, 21...Image acquisition unit, 22...Pseudo-fluoroscopy X-ray image creation unit, 23...Region of interest drawing unit, 24...Second region of interest creation unit, 25...Image matching unit, 26...Image display unit, 27...Control unit, 100...Pseudo-fluoroscopy X-ray image with region of interest (ROI) superimposed, 101...Pseudo-fluoroscopy X-ray image with 2D similarity image superimposed, 102... Image 101 shows an enlarged view of the square region 52, 200... Planned CT image, 201A, 201B... Virtual plane detector, 202A, 202B... Virtual X-ray source, 203A, 203B... (First) region of interest, 204A, 204B... Deformation of the region of interest considering the translational and rotational movement of bone, 205A, 205B... Second region of interest, 206A, 206B... Pseudo-fluoroscopic X-ray image, 207A, 207B... Fluoroscopy X-ray image, 300A, 300B... Region of interest cones, 301... 3D region of interest, 302... 3D region of interest after translational and rotational movement, 400A, 400B... (First) region of disinterest, 401A, 401B... Deformation of the region of disinterest considering the translational and rotational movement of bone, 402A, 402B... Second region of disinterest

Claims

1. A bed position control device for controlling the position of a bed on which a patient is placed, comprising: an image acquisition unit that acquires a fluoroscopic X-ray image of the patient taken by X-ray fluoroscopy; a pseudo-fluoroscopic X-ray image acquisition unit that acquires a pseudo-fluoroscopic X-ray image created from a three-dimensional image of the patient; a second region of interest creation unit that creates a second region of interest including the region of a target structure set in the pseudo-fluoroscopic X-ray image to correspond to the patient's positional displacement, from a first region of interest set in the pseudo-fluoroscopic X-ray image to include the region of a target structure set to determine the patient's position; and an image matching unit that compares an image in the pseudo-fluoroscopic X-ray image within the second region of interest with an image in the fluoroscopic X-ray image within the second region of interest, calculates the position of the bed necessary to match the patient's position in the fluoroscopic X-ray image, and moves the bed to the calculated position of the bed.

2. The bed position control device according to claim 1, wherein the second region of interest creation unit creates the second region of interest that encloses the first region of interest by deforming the first region of interest.

3. The bed position control device according to claim 2, wherein the second region of interest is created based on the extent of the region obtained by translating the first region of interest.

4. The bed position control device according to claim 2, wherein the second region of interest is created based on the extent of the region obtained by rotating the first region of interest.

5. The bed position control device according to claim 2, wherein the second region of interest is created based on both the expansion of the region obtained by translating the first region of interest and the expansion of the region obtained by rotating the first region of interest.

6. The bed position control device according to claim 1, wherein the image matching unit causes the second region of interest creation unit to recreate the second region of interest when the calculation accuracy of the bed position falls below a predetermined accuracy.

7. The bed position control device according to claim 1, wherein the second region of interest creation unit assumes the fluoroscopic X-ray image, defines the region formed by the overlap of the first region of interest and the X-ray source in three-dimensional space as a three-dimensional region of interest, deforms the three-dimensional region of interest in accordance with the patient's positional displacement, and creates the region formed by projecting the deformed three-dimensional region of interest onto a plane as the second region of interest.

8. The bed position control device according to claim 1, wherein the second region of interest creation unit creates the second region of interest on the day the patient is to receive radiation therapy.

9. The bed position control device according to claim 1, wherein the second region of interest creation unit creates and saves the second region of interest when creating a treatment plan for radiation therapy to the patient.

10. The bed position control device according to claim 1, further comprising a display unit that outputs a relationship between the first region of interest, the second region of interest, and the relationship between the position of the target structure in the fluoroscopic X-ray image and the position of the target structure in the pseudo-fluoroscopic X-ray image.

11. A method for controlling the position of a patient's bed using a bed position control device, comprising: acquiring a fluoroscopic X-ray image of the patient taken by X-ray fluoroscopy; acquiring a pseudo-fluoroscopic X-ray image created from a three-dimensional image of the patient; creating a second region of interest, including the region of the target structure, from a first region of interest set in the pseudo-fluoroscopic X-ray image to include the region of the target structure, corresponding to the patient's positional displacement; comparing the image in the second region of interest from the pseudo-fluoroscopic X-ray image with the image in the second region of interest from the fluoroscopic X-ray image; calculating the position of the bed necessary to match the patient's position in the fluoroscopic X-ray image; and moving the bed to the calculated bed position.

12. A radiotherapy apparatus comprising: a bed position control device according to any one of claims 1 to 10; and a radiation irradiation device for irradiating a patient on a bed whose position is controlled by the bed position control device with predetermined radiation.

Images