JP2019037795A - Image processing device, radiotherapeutic apparatus, and program - Google Patents

Abstract

To provide an image processing device capable of improving image quality and of image acquisition that suppresses influence of an area caused to appear by an object other than a subject in a short time.SOLUTION: A pixel group extending along an arrangement in an image matrix array is made to be a processing unit, and weight multiplied in common to all pixel values as an object belonging to the processing unit every processing unit is calculated and determined so that, for example, standard deviation of the whole image becomes minimum. Calculation determination processing is repeatedly performed for each processing unit of the image matrix array in succession while changing positions of the processing unit, and weight every processing unit is determined. The weight every processing unit is stored in a correction table in association with information about a position of each corresponding processing unit, and a corrected image is obtained by applying the weight stored in the correction table to an image to be corrected. The weight stored in the correction table is applied to an image newly acquired each time at a prescribed frame rate with respect to a subject 1 as a correction multiplied coefficient.SELECTED DRAWING: Figure 6

Description

  The present invention relates to an image processing technique for improving the image quality of an image obtained by X-ray imaging of a subject.

  When an image obtained by imaging a subject (target for treatment, diagnosis, etc.) with an X-ray imaging apparatus is used as a medical image for treatment, diagnosis, etc., the accuracy of the treatment, diagnosis, etc. is affected by the image quality. For this reason, the need for image processing for image quality improvement is greater for applications that require higher accuracy, and the image quality improvement algorithm that can advantageously achieve image quality improvement in terms of time required for the processing, It is desirable to realize the image processing.

  For example, in recent years, in order to increase the accuracy of particle beam therapy and other radiation therapy, dose calculation and irradiation accuracy have been improved, and the sophistication has progressed more than before. Along with this, “image-guided radiotherapy (IGRT)” (hereinafter, also simply referred to as “IGRT”), in which patient (subject) in-vivo information is acquired from an X-ray imaging apparatus or X-ray CT and reflected in treatment. ) Is being introduced to clinical settings.

As an application in the medical field, as described later, for example, an X-ray image of a patient before treatment is taken, and from this X-ray image and a CT image used at the time of treatment planning (this is a reference position), There is also a technique for performing a process (patient positioning) for reducing patient position error.
A technique that directly observes the inside of a patient using a fluoroscopic image during treatment beam irradiation, performs treatment beam irradiation when the tumor comes to an appropriate position, and does not irradiate the treatment beam at other tumor positions. (X-ray fluoroscopic respiratory synchronization) is also available.
Here, both the X-ray still image and the X-ray fluoroscopic image capable of observing the movement are collectively referred to as an X-ray image (hereinafter, the X-ray image may be used in this sense). Shall.

Needless to say, since the position accuracy and the like are greatly influenced by the image quality, as described at the beginning, the above-described IGRT and the like are greatly influenced by the image quality of the medical image to be applied.
Regarding image quality improvement, various image quality improvement methods have been proposed in the field of medical images and the like.

  In general, for example, the amount of information of an X-ray image is 14 to 16 bit gradation, but it is reduced to 8 bit gradation in a display observed by a person. The ease of observation is adjusted by selecting an 8-bit gradation range according to the observation site. For this reason, there is a method of selectively normalizing the pixel value distribution in consideration of the distribution (histogram) of pixel values within a high gradation level included in the X-ray image (for example, Patent Document 1, Patent Document 2, etc.).

In medical image processing, there are also examples in which an image processing method that discloses a technique for contrast normalization and a display method that aims to reduce stripe noise components (moire) in an image have been proposed (for example, Patent Document 3, Patent Document 4, Patent Document 5, etc.).
Note that the exemplified documents are referred to later in the description of the comparison with the image processing technique according to the present invention, which will be described later.

International Publication WO2010 / 035519 International Publication No. WO2010 / 074179 Japanese Patent No. 4913606 ([0137] to [0140]) Japanese Patent No. 4919408 ([0101] to [0104]) JP 2007-202916 A

  By the way, as already mentioned at the beginning, when improving the image for a medical image, not only can the image quality be improved, but also in terms of processing time (calculation time) required for the process for improving the image quality. However, it is desirable that this can be advantageously realized and effective image quality improvement can be achieved in a short time.

For example, in the case of a radiotherapy apparatus, it is necessary to improve the image quality in order to improve treatment results, but further, a radiotherapy apparatus that acquires a tumor position in real time using an X-ray fluoroscopic image (X-ray respiratory synchronization) Then, it is necessary to acquire about 15 to 30 images per second, calculate a tumor position, and determine treatment beam irradiation.
Therefore, when trying to improve an image for an X-ray image used in such a radiotherapy apparatus, the time required for the process for improving the image quality so as to meet the above-mentioned request simultaneously, It is required to perform image processing for improving image quality in a short time.

  On the other hand, when an X-ray image is used for treatment of a patient, image diagnosis, etc., an object other than the patient, such as a patient and a patient top plate, a fixture, is imaged (imaging) on the X-ray image. The patient (image) overlaps with the patient (image). For this reason, patient internal information and patient form information necessary for radiation therapy or the like may not be sufficiently obtained. In particular, since the treatment top has a portion containing an X-ray highly attenuated substance, this region is observed conspicuously on the X-ray image.

This will be further described with reference to FIG. 14. FIG. 14 (a) is a diagram which is also referred to in the description of the embodiment according to the present invention described later, and is a chest of a lung tumor patient lying on a top plate. X-ray image. As shown in the figure (a), in the image, there is a thick vertical line (a linear region part indicated by an arrow with a reference symbol Aa in the figure (a)) on the right side of the image, This is due to the X-ray highly attenuated substance on the treatment top plate, and is observed conspicuously.
Further, when focusing on the region, the position of the tumor as the treatment beam irradiation target (target, target) (the tumor position is indicated by being surrounded by white arrows in FIG. Although it is in the middle, as shown in the example, it overlaps with the treatment top plate (image) and the end of the tumor is unclear. As a result, in the case of such an X-ray image, when the tumor position is calculated based on the image as described above and the treatment beam irradiation is determined, the tumor position detection accuracy is lowered, and hence the treatment beam irradiation timing accuracy is reduced. It also becomes.

  Therefore, as shown in the above image example, there may be a case where patient internal information necessary for treatment, diagnosis, etc., patient form information, etc. cannot be sufficiently acquired due to the presence of an area caused by an object such as a top board other than the patient. For this reason, it is possible to delete, reduce, or reduce the area of an object area other than the patient when the image is improved by image processing, and it is possible to achieve image acquisition in a short time while suppressing the influence of the area. Is desirable.

Conventionally, depending on the method using the image histogram described above (Patent Document 1 and Patent Document 2), the image quality of the target image can be improved, but the image shown in FIG. When a medical image in which an object region other than the target subject (patient) exists as in the image example of (a) is targeted, the object (image region) other than the subject is reduced. I can't. Therefore, the above-mentioned method is used for real-time processing such as acquiring the above-described tumor position in real time, for example, in the case of X-ray fluoroscopic respiration synchronization, for the purpose of real-time understanding of the tumor position by respiratory movement. Corresponding to the improvement of image quality in line images, it is possible to easily respond to this, enabling high-speed image processing and achieving image acquisition in a short time while suppressing the influence of object areas other than the subject. Can not.
This is the same in the case of Patent Document 3 and Patent Document 4.

  Patent Document 5 tries to deal with the problem caused by a thin striped pattern having a constant period, with the problem of attenuation of the striped noise component (moire). Therefore, this document does not specify the focus on the above consideration. Moreover, in the case of this document, real-time processing is difficult with the calculation method (thus, when real-time processing is required, it is difficult to realize real-time processing that can easily respond to this and enable high-speed image processing).

Note that a technique of selectively deleting an image region (for example, extracting an image as a pattern) for a top plate as an object other than a patient is known.
However, this conventional method requires a long calculation time due to the large number of calculation steps. In addition, since the top plate position is different for each patient and for each treatment, as a result, in such a case, it is necessary to perform the above calculation every time, for each patient, for each treatment, etc., and the top plate position is determined and deleted in advance. It is difficult. For this reason, for example, it is difficult to perform image processing in a short time on an X-ray fluoroscopic image acquired in real time, and it is essentially unsuitable as a high-speed image quality improvement method.

Considering the above, the conventional image processing technique has a problem to be further improved when attempting to perform image processing for realizing effective image quality improvement at high speed.
Therefore, it is desirable to improve the image quality of an image obtained by X-ray imaging of the subject, and to improve the image quality, an object caused by an object other than the target subject itself is desirable. Image acquisition that suppresses the influence of the area can be realized in a short time.

  More desirable is the processing time (calculation time) required to improve the image quality of the entire image, which can be realized advantageously, and can improve the image quality effectively in a short time, and real-time processing is required. Even in the case of applications, it is possible to easily respond to this and to enable high-speed image processing.

  The present invention aims to provide an image processing technique that can be improved from these points based on the above considerations and also based on the considerations described later.

The present invention provides the following image processing techniques.
That is,
An apparatus for image processing having a correction function for improving the image quality of an image obtained by X-ray imaging of a subject,
One or more of the subject, a treatment table on which the subject lies, a cover of an irradiation device that irradiates the subject with a treatment beam, or a rotating bed of a gantry treatment room in which the treatment table is disposed Means for continuously acquiring images obtained by X-ray imaging of an object other than a subject;
Based on the image obtained by the means, a group of pixels extending along the arrangement in the image matrix array is used as a unit of processing, and for each processing unit, common to all pixel values to be a target belonging to the processing unit. A means for calculating and determining a weight to be multiplied, and as an index for determining the weight, the standard deviation of the entire image is minimized, or any one of variance, average value, entropy, and root mean square is used. Means for calculating and determining weights, and repeatedly executing the calculation determination processing for each processing unit of the image matrix array sequentially while changing the position of the processing unit. When,
Means for storing a weight for each processing unit determined by the means in a correction table in association with information on a position of each corresponding processing unit;
Means for obtaining a corrected image by applying the weight stored in the correction table by the means to the image to be corrected;
The means for obtaining the corrected image applies the weight as a correction multiplication coefficient to an image sequentially newly acquired at a predetermined frame rate for the subject after the weight is stored in a correction table. An image processing apparatus.

  Here, in the image correction by the means for obtaining the corrected image, it is checked whether or not image reacquisition by the means for obtaining the image is necessary. It is possible to further include means for newly performing the weight calculation determination processing based on the reacquired image.

  Further, the image reacquisition can be performed when the treatment table moves or when the position of the subject on the treatment table deviates beyond a preset state.

  Further, as a result of X-ray imaging of the subject, if the region imaged due to an object other than the subject is not oriented in the direction along the array in the image matrix array, the region is arranged in the image matrix array. Means for adjusting the directionality may be further provided along the direction.

  The means for adjusting the directionality can adjust the directionality by rotating the pre-correction image or defining coordinates for the object.

  In the calculation determination process in which the weight is determined for each processing unit by calculating and determining the weight for each processing unit, the start position of the processing is determined from the end of the image or an arbitrary process. Starting from the position of the unit, the weight calculation determination process can be executed in such a manner that the position is continuously changed to an adjacent processing unit.

  Further, in the means for determining the weight, the direction of the array as the processing unit can be the body axis direction of the subject.

The present invention also provides the image processing apparatus,
Radiotherapy beam irradiation means for irradiating a patient as a subject of radiotherapy with a radiotherapy beam,
Means for determining the start of treatment by irradiation with the radiation treatment beam;
On the condition of the start determination, image processing in the image processing apparatus and radiotherapy beam irradiation control in the radiotherapy beam irradiation means are executed, and the subject is obtained by X-ray imaging at a predetermined frame rate during radiotherapy. The image processing apparatus sequentially acquires images to be corrected based on the continuous X-ray fluoroscopy, the correction table is referred to by the image processing apparatus, and the images acquired sequentially by the continuous X-ray fluoroscopy Radiotherapy control by X-ray fluoroscopic respiration synchronized irradiation using the radiotherapy beam irradiation means using a corrected X-ray fluoroscopic image based on image processing by the image processing apparatus in real time Means for controlling the image processing device and the radiation therapy beam irradiation means to perform:
A radiotherapy apparatus comprising:

  The present invention is also a program for causing a computer to realize the image processing apparatus or the radiotherapy apparatus.

  According to the present invention, an improvement in image quality can be realized, and at the time of improving the image quality, an image acquisition can be realized in a short time while suppressing the influence of a region caused by an object other than the subject. Therefore, even in applications where real-time processing is required, it is possible to easily respond to this and enable high-speed image processing.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram for explaining an image processing apparatus according to an embodiment of the present invention and schematically showing an example of an applicable radiotherapy system (radiotherapy apparatus). The external view which shows an example of the inside of an example of an applicable irradiation room (treatment room) while showing an example of applicable X-ray imaging apparatus. The figure which shows an example of a structure of the control apparatus which can be applied. 6 is a flowchart illustrating an example of X-ray image processing for explaining the embodiment. Similarly, the flowchart which shows an example of the content of image processing. Similarly, the flowchart (correction table creation flow) which shows an example of the content of correction table creation. Similarly, the flowchart (image processing adaptation flow) which shows an example of the content of correction application. Similarly, a schematic diagram used for explaining an example of an image matrix arrangement. FIG. Similarly, explanatory diagrams of pixels and the like in an image matrix array. Similarly, a diagram for explaining an example of a process of calculating and determining weights. Similarly, the figure which uses for description of other examples, such as the process of calculating and determining a weight. Similarly, the figure which shows an example of a structure of a correction table. Similarly, the figure which uses for description of the state etc. which applied the weight with respect to all the rows. Similarly, the X-ray image figure used for description of image processing application. Similarly, another X-ray image diagram for explanation of application of image processing. The figure where it uses for description of the other example in calculation calculation of a weight. The flowchart which shows an example of an X-ray image process for description of other embodiment of this invention. Similarly, a diagram for explaining another example of X-ray image processing. The flowchart which shows an example of an image process with which it uses for description of other embodiment of this invention. Similarly, the figure used for description of the embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 schematically shows a radiation (particle beam) treatment system (treatment apparatus) as an example of a medical system or medical apparatus to which an image processing apparatus (method) according to an embodiment of the present invention can be applied. FIG.
In FIG. 1, a system generally indicated by reference numeral 100 irradiates an irradiation target of a patient 1 (irradiated body) with an accelerated charged particle beam as a therapeutic treatment beam in this example. A radiotherapy apparatus (particle beam therapy apparatus) for performing treatment is provided.

  The system 100 includes an accelerator 10 that accelerates and emits a beam B of charged particles, a beam transport system 20 that transports the beam B of charged particles emitted from the accelerator 10, and charged particles that are transported by the beam transport system 20. The irradiation apparatus 30 which irradiates the irradiation object (target, target, tumor part) 1a as a to-be-irradiated part of the patient 1 and the control apparatus 40 are provided.

The system 100 further includes a treatment table control device 60 and an X-ray imaging apparatus (X-ray imaging system) 70.
The treatment table 62 to be controlled by the treatment table control device 60 includes a top plate 61 (treatment table) on which the patient 1 lies, a treatment table main body that movably supports the top plate 61, and the like. In addition to the X-ray imaging apparatus 70, it is installed in an irradiation room (treatment room, irradiation treatment room) schematically indicated by reference numeral 80 in FIG. Specific examples of suitable practical configurations relating to the installation of devices and the like in the irradiation chamber including the plate 61 will be described later in detail).

The X-ray imaging apparatus 70 is an X-ray tube Xrt as an X-ray source for generating X-rays, and X-rays from the X-ray tube Xrt (schematic in FIG. 1). And a configuration having a dynamic flat panel detector DFPD as an X-ray flat panel detector for generating an image, which is a means for detecting a broken line arrow A: X-ray intensity after transmission through a patient) can do.
The X-ray tube Xrt and the flat panel detector DFPD are arranged to face each other through the patient 1 lying on the treatment table 62 (top plate 61) in the irradiation chamber 80, and at the time of imaging, the patient 1 (X-ray imaging) The two-dimensional X-ray image as a medical image obtained by photographing the in-vivo information and the in-vivo morphological information (including the patient as a subject for which the tumor position and patient position information are included). For digital image acquisition.

FIG. 2 is an external view showing a specific example of the applicable X-ray imaging apparatus 70 and the interior of a specific example of an applicable irradiation room (treatment room, irradiation treatment room) 80.
In this example, in addition to the component part of the X-ray imaging apparatus 70, an example of a specific state such as installation of components such as the component part of the irradiation apparatus 30 and the treatment table 62 component part is also shown. .
In this example, the irradiation device 30 for treatment beam irradiation is provided so as to be suspended from above the inside of the chamber 80 and covers and covers the irradiation device structure and the like in the chamber 80 so as not to be directly exposed. It has the structure provided with the irradiation apparatus cover 31 for doing.
The treatment table 62 can be configured to be capable of multi-axis control. Then, the required positioning using the X-ray image relating to the patient 1 (irradiation target 1a) is performed under the control of the treatment table control device 60 with respect to the couch top 61 serving as a patient couch. Can be controlled.

The X-ray imaging apparatus 70 has the following configuration in the example shown in FIG.
In this example, for two-way X-ray imaging, there are two sets of X-ray imaging systems (X-rays), a set of X-ray tube Xrt1 and flat panel detector DFPD1 and a set of X-ray tube Xrt2 and flat panel detector DFPD2. An imaging system and a fluoroscopic system) are used. The X-ray tubes Xrt1 and Xrt2 and the flat panel detectors DFPD1 and DFPD2 are provided so as to face each other so that these X-ray imaging directions intersect at an isocenter portion.
The X-ray tubes Xrt1 and Xrt2 are provided in the lower part of the chamber 80, and the flat panel detectors DFPD1 and DFPD2 are provided in the upper part of the chamber 80. Specifically, in the vicinity of the irradiation device 30, preferably as shown in FIG. 2, in the irradiation device cover 31 portion suspended from above, it is attached and installed with an integral structure on the lower edge of the suspended portion. It can be of a configuration. Note that the illustrated configuration is an example, and the applicable X-ray imaging system component portion can be installed at any position, and the present invention does not preclude such implementation.

  In the imaging system according to this example, when X-rays transmitted through the patient 1 by X-ray imaging are used, and this is detected and imaged to obtain an X-ray image of the patient 1, the patient before the treatment is treated before the treatment. When X-rays are taken to acquire an image to be used for patient positioning by treatment table control, and during treatment beam irradiation control of the irradiation device 30, for example, at an image frame rate of about 15 to 30 images per second When calculating the position of the irradiation target (tumor part) 1a and determining the treatment beam irradiation (treatment beam irradiation timing), when acquiring a frame image used for X-ray (fluoroscopic) type respiratory synchronization control, each image is acquired. At the time of acquisition, X-ray images in two directions can be acquired simultaneously by the two sets of imaging systems.

  When images are acquired in two or more X-ray imaging directions, the tumor position and patient position can be calculated with three translational axes and three rotational axes.

  Note that the image processing according to the present invention can also be applied to one-way X-ray imaging. In this case, one or more X-ray images may be acquired from X-ray imaging in one direction.

Returning to FIG. 1, the control device 40 controls the entire system 100 including the control for the accelerator 10 and the beam transport system 20 as well as the control for the irradiation device 30, the treatment table control device 60 and the X-ray imaging device 70 described above. It is illustrated as an example.
Although the control device 40 is illustrated as one control system in the figure, distributed processing includes accelerator control, beam irradiation control, image treatment control, and other controls (including treatment table control) executed in the system 100. In the case of executing by the above, it may be configured to have a plurality of computers. Here, as is known, a computer (computer), as is known, a central processing unit (CPU or the like) which is a main control means for controlling operations such as calculation and processing of data and program execution, a memory for storing data and programs It comprises a memory as a device, an input / output device, and the like.

FIG. 3 is a diagram illustrating an example of the configuration of the control device 40 as a functional block (functional unit).
In the illustrated example, the control device 40 includes an arithmetic processing unit 41, an image processing control unit 42 for X-ray image processing control showing details of the contents together with a program flowchart and the like described later, and a beam for treatment beam irradiation control. The irradiation control unit 43, the accelerator control unit 44 that controls the accelerator 10 and the beam transport system 20, an input unit 45, an output unit 46, and a storage unit 47 may be included.

In addition, although not shown in figure, the control apparatus 40 shall perform data transmission / reception, communication, etc. with the database which stored the information etc. which were acquired at the time of a treatment plan in a treatment plan system, and other external systems. In relation to image processing, for example, CT image data used at the time of treatment planning, which is used for positioning control performed by the treatment table control device 60, is also input.
Similarly, in relation to image processing, the input unit 45 and the output unit 46 include, for example, an X-ray imaging instruction and other commands related to the X-ray tube Xrt (Xrt1, Xrt2) of the X-ray imaging apparatus 70. Used for input of In the output unit 46, a display function for displaying a screen of image data (including a corrected image subjected to image processing for correction by the X-ray image processing control) on the screen, It may be used for data transmission to other systems using data.

  In the present system example, the beam irradiation control unit 43 for the irradiation apparatus 30 similarly receives an X-ray image from the flat panel detector DFPD (DFPD1, DFPD2) of the X-ray imaging apparatus 70 in relation to image processing. It is assumed that treatment beam irradiation control is executed using image data supplied as input information so that treatment by the used IGTR can be performed.

From the viewpoint of real-time properties in radiotherapy, IGRT “image-guided radiotherapy” is used as an example in which treatment is performed by irradiating a treatment beam (B) from the treatment beam irradiation device (30) to the irradiated site (1a). ”, The IGRT performs treatment while confirming time-dependent changes in the patient's body (eg, time-dependent changes including physiological phenomena such as breathing, heartbeat, and swallowing) using X-ray images and X-ray CT images. If so, it can help improve treatment outcomes.
Here, particularly in the treatment of the trunk, the position of the tumor (irradiated site, target) and organ changes over time due to respiratory movement. For this reason, if it is not possible to cope with such a situation, the target tumor position as the irradiated site may be out of the treatment beam irradiation field. For the purpose of enabling irradiation control corresponding to such position fluctuations, it is necessary to grasp the tumor position in the patient in real time, and an IGRT device using an X-ray image should also be introduced and used in the clinical field. The effect can be exhibited.

  Therefore, from this point of view, in the present system 100, when the treatment of the trunk of the patient 1 is targeted, the inside of the patient 1 is directly used using the X-ray fluoroscopic image during the irradiation control of the treatment beam B by the irradiation device 30. Irradiation control (X-rays), which is controlled so that the treatment beam B is irradiated when the tumor (1a) comes to an appropriate position, and the treatment beam B is not irradiated at other tumor positions. Such control is executed by the beam irradiation control unit 43 based on (perspective) type respiratory synchronization). As a result, exposure of the patient 1 to normal tissue can be reduced, and the treatment beam B can be intensively irradiated only to the tumor (1a), leading to higher treatment accuracy.

  Further, in parallel with the treatment beam irradiation control, the system 100 uses the image processing control unit 42 to improve the image quality of the X-ray imaging apparatus 70 at a high speed, thereby improving the algorithm (X-rays) useful for improving the treatment accuracy of IGRT. Image processing is performed so that control by the image quality improvement algorithm of the photographing apparatus becomes possible.

  Specifically, the image processing control unit 42 performs X-ray imaging and flat panel detector DFPD (DFPD1, DFPD2) data on the X-ray imaging device 70 under the overall control of the control device 40 (arithmetic processing unit 41). In the case of continuous imaging of X-ray fluoroscopy during treatment by IGRT, a predetermined frame rate applied to the continuous imaging is assumed to be read and transferred to a computer (X-ray image data transfer process). It is assumed that the control necessary for the image processing control system is executed so that the above control can be performed in response to (for example, an image collection frame rate obtained by acquiring about 15 to 30 images per second).

Further, preferably, in the present system 100, the storage unit (memory) 47 of the control device 40 realizes image quality improvement at high speed by adapting image processing to X-ray images in real time together with various programs. An X-ray image processing program for performing an image correction process so as to reduce the influence of an object (image) other than the patient 1 on the image is stored in advance, and the image processing control unit 42 acquires in accordance with the program. Image processing is executed based on the X-ray image information and the like.
The storage unit 47 stores and stores a correction table 47A (see FIG. 12) that is created in the program and applied to the correction target image, and controls that are used during image processing. It is also used for storing data such as variables and functions. Here, the variable includes a weight (w) (weighting coefficient) used for correction for improving the image quality of an image, which will be described later, and stored in the correction table.

The image processing according to the present embodiment will be further described with reference to FIG.
4 to 7 are diagrams showing examples of procedures and the like provided for explaining the image quality improvement correction processing according to the present embodiment, and FIG. 8 is a two-dimensional xy plane also used for explanation. FIG. 9 is a diagram schematically illustrating an example of an image matrix array, and FIG. 9 is a diagram illustrating pixels at each pixel position (x, y) constituting the image in the matrix array, which is also provided for explanation. 10 to 11 and 13 are diagrams for explaining an example of a weight (w) calculation determination process in image processing and an example of an image after correction (after image quality improvement processing), and FIG. 12 is a correction table as described above. It is a figure which shows an example of 47A.

  FIG. 4 shows a flowchart according to an example of an X-ray image processing program. In this program example, X-ray imaging (step S10), image processing (step S20), and output (step S30) are respectively performed. The following basic process is shown.

  In step S10, the patient 1 lying on the treatment table 62 (top plate 61) is X-rayed by the X-ray imaging apparatus 70, and the X-ray captured by the image processing control unit 42 by the flat panel detector DFPD (DFPD1, DFPD2). Get an image.

  In step S20, the image processing control unit 42 acquires a corrected image that has been subjected to image processing by the following processing. In this example, step S20 includes a process for determining the weight (w) and storing it in the correction table 47A, and an image correction process.

In the weight determination process, an acquired image is used as an original image, and based on this, a pixel group extending along the array in the image matrix array (FIG. 8, FIG. 9, etc.) is used as a processing unit. , Common to all target pixel values (Im (k, 1), Im (k, 2),..., Im (k, m-1), Im (k, m)) in the processing unit The weight (w) to be multiplied is calculated and determined. In this case, in determining the weight (w), in this example, the weight (w) is calculated and determined so that the standard deviation of the entire image is minimized. For each processing unit in the image matrix array, the processing unit is repeatedly executed while changing the position of the processing unit to determine the weight (w) for each processing unit.
The weight (w) for each processing unit determined as described above is expressed as w (1), w (2),..., W (m−1) as in the data configuration illustrated in FIG. ) And w (m) are stored in association with (corresponding to) the corresponding position information of each processing unit, thereby creating the correction table 47A.
In the image correction process, a corrected image is obtained by applying the weight (w) of the correction table 47A determined and stored in this way to the correction target image as illustrated in FIG.

  Thus, it is possible to improve the image quality of the entire image and to achieve image acquisition in a short time while suppressing the influence of an object area other than the patient 1 (for example, the area Az (FIGS. 9 to 11 etc.) such as the top board). An example of a specific procedure and a detailed program example are shown below.

  Note that in step S30, the corrected image whose image quality has been improved as described above is used for screen display on a display for, for example, treatment or diagnosis, and other image data is used. Processing such as sending to the beam irradiation control unit 43 for radiotherapy control by X-ray (fluoroscopic) respiratory synchronized irradiation as a control system or the like can be executed.

  FIG. 5 is a flowchart showing an image processing program as an example of the contents of the image processing (step S20) in the X-ray image processing flowchart (FIG. 4). In this program example, correction table creation (first processing) (step S21) and correction application (second process) (step S22).

An example of the contents related to step S21 and step S22 will be described below as (procedure: 1-1) to (procedure: 1-5). The specific contents are shown as an example of a flowchart. 6 and the flow shown in FIG.
FIG. 6 is a flowchart (correction table creation flow) of a correction table creation program showing a specific example of the contents of the correction table creation (first processing) corresponding to step S21 in the image processing flowchart (FIG. 5). 7 is a flowchart (image processing adaptation flow) showing a specific example of the contents of correction application (second processing) corresponding to step S22 in the image processing flowchart (FIG. 5).

  In the case of the example program shown in FIG. 6, the program includes the following steps S211 to S219.

<Step S211>
Reading flat panel detector DFPD image (Im):
“Import DFPD image (Im)”

<Step S212>
Initialization of weight (weighting coefficient) (w (x) = 1, x = 1,..., Column number):
“Initialize weighting factor (w (x) = 1, x = 1,..., Column number)”

<Step S213>
Change the weight (w (k)) of the kth column:
“Change weighting factor at kth column (w (k))”

<Step S214>
(Image) Apply weight w (k) to Im (Im ′ = Im × w (k)):
“Apply the weighting factor to Im (Im ′ = Im × w (k))”

<Step S215>
After n iterations, calculate the standard deviation in (Im × w (k) image) Im ′ SD (n):
“Calculate standard deviation within Im'after nth iteration SD (n)”

<Step S216>
Is the standard deviation SD (n) the minimum:
“SD (n) is lowest?”

<Step S217>
Whether all columns end:
“All column are complete?”

<Step S218>
Shift to next column (k = k + 1) (for column, increment k by the value 1):
“Shift next column (k = k + 1)”

<Step S219>
Processing Exit:
“Complete”

  In the case of the program example of FIG. 7, the following steps S221 to S223 are included.

<Step S211>
Reading flat panel detector DFPD image (Im):
“Import DFPD image (Im)”

<Step S222>
Apply weight to image (Im):
“Apply the weighting factor to (Im)”

<Step S223>
Processing Exit:
“Complete”

  Referring also to FIG. 5, FIG. 6, and FIG. 7, the procedures relating to step S21 (first process) and step S22 (second process) are defined as (procedure: 1-1) to (procedure: 1-5). The contents will be described below.

(Procedure: 1-1)
An X-ray image of patient 1 is acquired.
As described above, the X-ray imaging directions by the X-ray imaging apparatus (X-ray imaging system) may be two or more directions or may be one-way X-ray imaging.
This X-ray image is transferred to the computer, and image processing after (Procedure: 1-2) is performed.

(Procedure: 1-2)
Next, the weight (w) is initialized. The initialization process corresponds to step S212 in FIG. 6 (w (x) = 1).
FIG. 8 is a diagram for explaining image processing by creating a correction table and the like, schematically showing an example of an image matrix arrangement in which pixels are arranged in a matrix of “m rows × m columns” on the xy plane. FIG. 9 is a diagram for explaining image pixel values and the like at each pixel position before the start of image processing.
As shown in these figures, the pixel (pixel value) at the image pixel position (x, y) is indicated as Im (x, y) (the same figures). The y direction is the patient's head-to-tail direction.

Here, the weight w (x) is obtained and stored as a final value in the correction table 47A used for correction application in step S22. , For each column, a variable to multiply (multiply) all pixel values in each column. At the start of image processing, as a result of the initialization described above, all the pixels (a group of column pixels “(Im (1,1) to Im (1) consisting of a plurality of pixels extending in the array y direction in the image matrix array) 1, m)), ..., (Im (k, 1) to Im (k, m)), ..., (Im (m, 1) to Im (m, m)) " The variable has the value 1 (w (x) = 1).
That is, in the initial state, the image that is obtained by multiplying w (x) of 1 as the initial value is the same as the original image (read original image).

Further, in FIG. 9, a portion indicated by reference symbol Az (a group of a plurality of pixel portions along the image y direction surrounded by a broken line in the drawing) is an object other than the patient 1 in improving the image quality. Represents a region such as a top plate that is to be deleted, reduced or reduced (region caused by X-ray highly attenuated substance). Therefore, this figure shows an object to be reduced in the image (original image to be corrected). An example of a state in which a region Az such as a top plate is also imaged and included is also shown.
In FIG. 9, from the point of view of simplification, only the column pixel group for one vertical column is displayed as the region Az portion on the image matrix arrangement. For example, FIG. In fact, such an area Az portion includes a plurality of continuous (adjacent) columns (in the x direction, a plurality of pixels), like an image portion (thick vertical line portion) indicated by an arrow Aa in the X-ray image example of FIG. For example, it may appear as a single-like (single) linear region.
Further, these points of simplification of illustration and the like are also expressed in the same manner in FIG.

(Procedure: 1-3)
Next, using the column pixel as a processing unit as described above, for example, from the X-ray image end (x = 1) (the left end of the x coordinate in FIG. 9), the position (column position) where weight calculation starts is the X-ray. The weight w (x) to be obtained is sequentially calculated and determined (as the image edge).
First, the weight (w (1)) in the first column in the image y direction is changed from the initial value (= value 1), and the standard deviation of the entire image is calculated each time. Then, the weight (w (1)) is changed until the standard deviation value becomes the smallest. In order to efficiently obtain the weight at this time, the final value w (1) can be obtained in a short time by using, for example, optimization calculation.
In this way, the weight w (1) of the first column (x = 1) in the image y direction is determined.
The above processing corresponds to a loop of steps S213-S214-S215-S216-S213 in FIG.

(Procedure: 1-4)
Next, using the image obtained by multiplying the first pixel value Im (1, y) in the image y direction (x = 1) by the final value of w (1), the second column (x = 2) in the image y direction is used. The weight (w (2)) is determined.
FIG. 10 shows an example of the process of calculating and determining the weight (w (1)) relating to the column in the image y direction, in the illustrated example, the first column. In the column pixel Im (1, y), FIG. , W (1) The example of data when the final value is multiplied as a common multiplication coefficient is “Im (1,1) × w (1), Im (1,2) × w (1),. .., Im (1, m-1) .times.w (1), Im (1, m) .times.w (1) "is also shown for corresponding to each pixel position in the drawing.
(Procedure: 1-3) to (Procedure: 1-4) are repeated while changing the position of the x-th column in the image y direction by the above procedure.

FIG. 11 shows an example of the process of calculating and determining the weight (w (2)) relating to the column in the image y direction, in the illustrated example, the second column in such an iterative execution process. 2, y) is multiplied by w (2) final value as a common multiplication coefficient. Examples of data are “Im (2,1) × w (2), Im (2,2) × w (2 ),..., Im (2, m-1) .times.w (2), Im (2, m) .times.w (2) "in order to show the corresponding pixel positions in the drawing. It is also a thing.
In this way, the weight calculation / determination process is sequentially executed for all the columns in the image y direction, and the correction table 47A is created by performing the above-described processing for all the columns.
In FIG. 6, the process corresponds to the loop of steps S213-S214-S215-S216-S213 described above and the loop of steps S213-S214-S215-S216-S217-S218-S213 including the loop.
Specifically, in step S217 of FIG. 6, every time this step is executed, it is checked whether or not weight calculation determination processing has been executed for all columns, and weight calculation determination processing has been executed for all columns. Then, the correction table creation is completed.

In the weight calculation / determination, the processing contents when the standard deviation minimization of the entire image is used can be expressed as follows.
Im (x, y) × w (x), w (x) = min (std (Im (x, y) × w (x))) (Equation 1)
The above std () means a function for calculating a standard deviation for “Im (x, y) × w (x)”, and min () calculates a minimum value for the function “std ()”. Means a function.

In the correction table creation (first process) in step S21 in FIG. 5, the process is executed in the image processing control unit 42 (control device 40) according to the above-described procedure. Therefore, by the irradiation device 30 (the control device 40, the beam irradiation control unit 43),
Prepare before treatment beam irradiation.
Thus, the weights w (1), w (2),..., W (m−1), w (m) for each column in the image y direction are determined, and the weight w (x) for each column to be determined. Is associated with (associated with) information indicating a column (x coordinate data (1, 2,..., K,..., M−1, m) of a column position) as shown in FIG. The data is stored in the correction table 47A of the storage unit 47 (control device 40).

(Procedure: 1-5)
The weight w (x) stored in the correction table 47A can be suitably used to obtain a corrected image by applying it to the image to be corrected.
That is, in the correction application (second process) in step S22 of FIG. 5, the image processing control unit 42 (control device 40) refers to the correction table 47A and sets the weight w (x) to the correction target image (for example, a predetermined image). Applies to images acquired at frame rate.
Such correction application processing corresponds to steps S221 to S222 in FIG.

Specifically, one-way or two-way X-ray images newly acquired sequentially at the time of treatment by the irradiation device 30 (the control device 40, the beam irradiation control unit 43) are transferred to the computer within a few milliseconds. When acquiring X-ray images in two directions, it is possible to acquire images in two directions one by one or simultaneously. However, when the X-ray image is transferred to the computer, the calculated weight (w (x)) ) On the input image, the image quality can be improved in a few milliseconds, and real-time performance is possible (FIG. 7).
In this case, the image processing control unit 42 does not need to newly perform recalculation (calculation) processing such as the above (procedure: 1-3) to (procedure: 1-4), and stores it in the storage unit 47. Using the calculated weight (w (x)) data read from the correction table 47A and applying this data, image processing can be executed, and the above is made possible.

FIG. 13 shows the calculated weights w (1), w (2), w (3),..., W (k)... W (m-1), w ( A corrected image of the result of applying (multiplying) m) to each column pixel of all corresponding columns in the matrix arrangement in the image is shown.
As shown in the figure, the first column pixel in the image y direction (x = 1), the second column pixel in the y direction (x = 2), the third column pixel in the y direction (x = 3),. The k-th (x = k) column pixel in the y direction,..., The m−1th (x = m−1) column pixel in the y direction, and the mth (x = m) column in the y direction. For each pixel, the corresponding calculated weights w (1), w (2), w (3),..., W (k). , W (m) are respectively applied as multiplication correction coefficients, and as a result, the corresponding correction is applied to all “m × m” pixels in the entire image. It is shown that the image quality is improved by applying a coefficient (weight).
FIG. 13 also shows that the area Az of the top plate or the like has been reduced or reduced as a result of the above-described improvement in image quality as compared with FIGS. 9 to 11 (in the state of the one-dot chain line in FIG. 13). Also shown). Further, FIG. 13 shows a state in which even if the region Az is a single linear region (one) in the image, it is reduced in the corrected image.

By applying the above image processing to the X-ray image in real time by the above method, the image quality can be improved at high speed, and the influence of an object (image region Az) other than the patient 1 on the image can be reduced.
Therefore, for example, the present invention is suitable for use in improving image quality in an X-ray image (X-ray fluoroscopic image) for the purpose of grasping a tumor position in real time by respiratory movement.

FIG. 14 has already been mentioned in the discussion at the beginning. Referring to this again, this figure is a chest X-ray image of the patient 1 lying on the top board (treatment top board) 61 for explaining the result of the above image processing application (correction application), (A) is a figure shown as an example of a comparative image (image before the said image processing implementation), The figure (b) is the image after the process obtained by implementing the said image processing shown in contrast with this (b). It is a figure of an example.
Comparing the image before image processing in FIG. 6A with the image after image processing in FIG. 4B, the image quality of the entire image is improved in the image after processing in FIG. .

Further, in the image of FIG. 6A, the X-ray of the treatment top 61 is a thick vertical line on the right side of the image (the region indicated by the arrow with the reference symbol Aa in FIG. 4A). Although it has already been described that it is due to a highly attenuated substance and can be observed largely and conspicuously in an X-ray image, this is also improved in the image after processing in FIG.
Furthermore, this X-ray image example is an example of a lung tumor patient, and in the image before the above image processing in FIG. (A), the white arrow at the tumor position (image, middle in the figure (a)) It has already been described that the region surrounded by () is overlapped with the treatment top plate 61 (image) and the end of the tumor is unclear, but also after this processing in FIG. In this image, since the treatment top plate area is reduced after processing, the end of the tumor can be easily recognized.
Therefore, as a result of obtaining such a corrected image, as described above, even when the tumor position is calculated based on the X-ray image and the treatment beam irradiation is determined, a decrease in the tumor position detection accuracy is avoided, and the treatment beam is avoided. It is also possible to improve the accuracy of irradiation timing.

FIG. 14 is also taken on an X-ray image as a result of an object other than the patient 1 (image region Az shown in FIG. 13), for example, due to the cover 31 of the irradiation apparatus 30 as shown in FIG. An example of an image in which the imaged image appears (at the same time as the treatment top plate region) is also shown.
As described above, if the flat panel detector DFPD (DFPD1, DFPD2) in the X-ray imaging apparatus 70 can be installed as an integral part of the irradiation apparatus cover 31 portion, When trying to design and arrange a flat panel detector in the upper part (space), it is effective because it is not necessary to use a dedicated support structure for supporting the flat panel detector. On the other hand, in the X-ray image by the flat panel detector having such an arrangement, an image caused by the irradiation device cover 31 may be imaged.
In the image of FIG. 14A, a slightly thin white vertical line indicated by an arrow with a reference symbol Ab on the right side of the treatment top plate (image) portion is an example of such a shadow of the irradiation device cover 31. Yes, it appears on the image before the processing in FIG. 11A as a linear region, separately from the treatment top plate (image), but this also after the processing in FIG. The image has been improved.

FIG. 15 is an X-ray image diagram showing another example for explaining the result when the image processing (correction application) is applied, and the rotating bed in the gantry treatment room is projected on the X-ray image. This is an application example of image processing according to the present embodiment with respect to the X-ray image.
As described below, in the case of a gantry treatment room, in the same case as seen in the image example of FIG. 14, the object is an object other than the patient to be X-rayed (image area Az shown in FIG. 13). ) As a result of the X-ray superabsorbent part of the rotating bed sometimes overlapping the patient (image). In such a situation, the tumor position is calculated based on the X-ray image to determine the treatment beam irradiation. Similar to the treatment top plate (bed) region and the like, there is a possibility that the tumor position detection accuracy may be reduced.

  In the gantry treatment room, the irradiation apparatus rotates around the patient to enable treatment beam irradiation from various directions. However, depending on the installation layout of the X-ray imaging apparatus, the rotating floor region is projected on the X-ray image and overlapped with the patient to form an image, which may cause a decrease in tumor position detection accuracy. In addition, since a treatment beam irradiation angle changes for every patient, a correction parameter cannot be prepared beforehand.

Based on such consideration results, an example of applying the above image processing (correction application) to a correction target image (patient chest X-ray image) in which a rotating bed in a gantry treatment room is projected on an X-ray image This is shown in FIG.
In addition, the same figure also shows the example of an image which the area | region part of the above treatment top boards (beds), etc. have appeared simultaneously.

FIG. 15A is an image before image processing, and FIG. 15B is an image after image processing shown in contrast thereto. In the image of FIG. 15A, an arrow with reference symbol Ac is shown. The area | region part to show is a X-ray high absorption part (rotary bed X-ray high absorption area | region) of a rotation bed.
On the other hand, when the image processing according to the present embodiment is applied, it can be seen that the object region other than the patient including such a rotating bed is reduced as in the image after the processing in FIG. As shown in FIG. 5B, the rotating bed area is also reduced (see the arrow portion with the symbol Ac), and the treatment couch (bed) area portion is also reduced.
Therefore, as a result of obtaining such a corrected image, a decrease in tumor position detection accuracy can be avoided as described above, and as a result, the accuracy of treatment beam irradiation timing can be improved.
The image processing according to the present embodiment can be applied to such a case to exhibit the effect.

In addition, if it mentions further about patent document 5 illustrated as a prior art document here, as also mentioned at the beginning, the document 5 makes the subject the attenuation | damping of a striped noise component (moire).
Specifically, it is intended for the case of using a grid for removing scattered X-rays. Generally, a scattered radiation grid is a parallel one with a thickness interval of 1-2 mm using a metal such as lead. Or things arranged in a grid. Therefore, since it is difficult to recognize the grid unless the eyes are focused, the “striped noise component” as the attenuation target is composed of a fairly thin component.
Moire is interference fringes, and has a narrow pattern with a high period (repetition frequency) as described above. In addition, in order to generate moire, the fringe frequency shifts. There is a need to. That is, a certain large stripe pattern is required. In other words, no moire occurs with a single stripe. Therefore, from these facts, in the case of the document 5, it is intended to deal only with the problem due to the thin stripe pattern having a fixed period, and it can be said that only a method for reducing the moire of an X-ray image using a grid is disclosed. .

On the other hand, according to the image processing according to the present embodiment, as already described in connection with FIGS. 9 to 13 and the like, as the region Az portion imaged due to an object other than the patient 1, Even if the image appears as an independent single linear region on the screen, effective image processing for correction such as reduction is possible.
Furthermore, as shown in FIGS. 14 and 15 above, what is focused on as an object other than the patient 1 targeted for reduction, reduction, etc., is specifically the treatment top plate (bed) region and irradiation device cover as illustrated. Each of the area and the rotating bed area is individually and independently in various forms (for example, the width of the linear area depending on the properties of the X-ray highly attenuated substance, etc. (thick or slightly thin). In addition, there are special characteristics such as the possibility of being imaged in the image in the state where the positional relationship such as the interval of each region when the images are simultaneously imaged, etc. Even in this case, in the image processing according to the present embodiment, as described above, the image quality can be improved at high speed, and the influence of various objects other than the patient 1 on the image can be effectively achieved. What can be reduced There, therefore, are to be due to the document 5, another having different Viewpoints in this regard.

  Furthermore, in addition, in the same document 5, the stripe noise reduction is performed by the filter processing, but the image processing according to the present embodiment is not, and the same document 5 describes the tumor position due to the respiratory movement described above. As described above, there is no place for touching image quality improvement in X-ray images for the purpose of grasping in real time, and that real-time processing is difficult with the calculation method.

In view of the difficulty of performing image processing in a short time, it may be possible to use thinning (for example, Patent Document 3). That is, since it is difficult to perform image processing in a short time on an X-ray fluoroscopic image acquired in real time, therefore, it is conceivable to adapt the image processing by thinning out the acquired X-ray fluoroscopic image, for example. In order to determine treatment beam irradiation with higher accuracy, it is preferable to perform image processing on all acquired X-ray fluoroscopic images.
According to the image processing according to the present embodiment, even in such a case, since real-time property is possible as described above, it is possible to easily perform image processing on all acquired X-ray fluoroscopic images. In view of this, the image processing according to the present embodiment is suitable.

  Below, the modification etc. are further described about the said image processing according to this embodiment.

(First applied variation of this embodiment)
In the above embodiment, as described in (Procedure 1-3), the weight of each column is calculated sequentially from the X-ray image end (x = 1) (the left end of the x coordinate in FIG. 9). However, this may be started from an arbitrary position (an arbitrary column position).
In this case, however, it is necessary to calculate the weight continuously (for example, so as to shift the columns one column at a time (see S218 in FIG. 6)), and it is preferable to do so. If the weight of the jump position (the position of the jump row) is calculated, the expected result (as shown in FIGS. 13 to 15 can improve the image quality of the entire image, and An image processing result having a correction processing function that can reduce an area caused by an object other than the patient 1 and suppress an influence thereof in a short time may not be obtained.
For this reason, from this point of view, it is assumed that the correction function is equivalent to the correction function, and in order to ensure that it is always obtained, in the weight calculation determination process, the position is continuously changed with respect to the processing unit. Thus, it is preferable to calculate the weight.
This embodiment may be implemented in this way.

(Second applied variation of this embodiment)
In the above embodiment, in (Procedure: 1-2) or the like prior to (Procedure: 1-3), weight initialization is performed uniformly at that time point ((Procedure: 1-2), S212 in FIG. 6). Although the mode in which the value 1 is set as the initial value has been described, in order to further reduce the calculation time, the following processing contents as shown in an example in FIG. 16 may be used.

FIG. 16 is a diagram for explaining another example in the weight calculation determination process, and also shows a modification example regarding the initial value setting of the weight.
As shown in the figure, for example, from the X-ray image end (x = 1) (x coordinate left end in FIG. 9) side, the first row in the image y direction (x = 1), the second in the image y direction (x = 2), the third column in the image y direction (x = 3),..., The weight in the order of the k−1th column (x = k−1) in the image y direction. The calculation decision has been made, and now, for the weight w (k) in the k-th column in the image y direction, iterative procedure as described above ((procedure: 1-3), S213-S214-S215-S216 in FIG. It is assumed that the weight w (k) is calculated (calculated) and determined by the loop (S213).
In the case of the above aspect in which the value 1 is uniformly set as the weight initial value, the weight w (k) is also changed from the initial value 1 to the weight value as a variable to obtain the correction function described above. Thus, the final value of the weight w (k) to be obtained is determined so as to minimize the standard deviation of the entire image.
On the other hand, in order to further improve the calculation time, in order to shorten the calculation time, the processing unit to be calculated (currently, it is to be calculated in a processing loop based on an iterative procedure for weight calculation determination processing) The weight value already determined in the processing for the previous processing unit (adjacent position column) (the weight determination value regarding the corresponding adjacent processing unit that has been calculated) can be used effectively.
This applied variation is based on such an idea.

  In order to shorten the calculation time, the weight (w (k-1) or w (k + 1)) determined at the position (column position) immediately before (w (k)) to be calculated. When the calculation (weight calculation determination) is performed with the value of as the initial value (= value w (k−1)) for the weight w (k) to be calculated and determined this time, the calculation process is shortened.

Thus, compared with the case where the initial value of the weight is set to the value 1, the weight value is changed according to the above (procedure: 1-3), the weight to be obtained is calculated, and the final value is determined. Already calculated for adjacent processing units that are close to each other in terms of pixel values in the image, as long as there is no need to perform uniform processing such as always having to start changing from a fixed value of 1. Since the weight value of each can be used effectively while changing the position of the processing unit sequentially (procedure: 1-4), the calculation time can be further shortened. This contributes to shortening the time required for processing and is effective.
This embodiment may be implemented in this way.

(Third applied variation of this embodiment)
In the above embodiment, as exemplified in (Procedure: 1-3), (Procedure: 1-4), etc., in the weight calculation determination process, the weight is changed until the standard deviation value of the entire image is minimized. Trying to determine the final value.

  Here, the minimization of the standard deviation of the entire image is used as one index for determining the weight. However, any index may be used as long as the weight that reduces the object area other than the patient can be calculated. For example, variance, average value, entropy, and root mean square can be used.

  In addition, each said application deformation | transformation form can be implemented in the form which is respectively independent, or combines 2 or more.

Next, an image processing apparatus (method) according to another embodiment (second embodiment) of the present invention will be described with reference to FIG.
FIG. 17 shows an image processing apparatus (method) in the present embodiment. Under predetermined conditions, image reacquisition and weight (w (x)) calculation based on this (correction table (47A) recreation processing) ) Is a program flowchart showing an example of the contents of X-ray image processing when necessary processing is further added to the embodiment (first embodiment).
In addition, this embodiment is also a modification of the first embodiment and each application modification thereof. Therefore, detailed description of the same or similar components as in the first embodiment will be omitted.

  The main part of the present embodiment will be described below as (procedure: 2-1) to (procedure: 2-2), and the processing content will be described with reference to the flowchart example shown in FIG.

In the case of the program example of FIG. 17, the process includes steps S410 to S450.
In FIG. 4, step S410 (X-ray imaging: obtaining an X-ray image of a patient lying on the treatment table 62 (top plate 61)) may be step S10 of FIG. 4 in the first embodiment, and step S420. The (weight (w (x)) calculation, (correction table (47A) creation)) may be equivalent to step S21 in FIG. 5 (corresponding to the flow in FIG. 6) in the first embodiment.
Step S430 (refer to correction table (47A) (image correction)) may correspond to step S22 in FIG. 5 and the flow in FIG. 7 in the first embodiment.

Furthermore, this program example includes step S440 for determining whether or not there is movement control of the treatment table 62 (top plate 61), and step S450 for determining the end, and these processes are described in the first embodiment. Has been added.
The added step S440, for example, after the process has been advanced from step S420 to step S430, during the execution of the loop process of step S430-step S440-step S450-step S430, every time the step S440 is executed. It is possible to check (monitor, monitor) whether or not image re-acquisition (step S410) and new weight calculation based on this (step S420) is necessary. By providing, if the determination result is affirmative (Yes), the X-ray image processing control can be executed so that the process can be switched to return to step S410.

Further, the added step S450 determines whether or not to terminate the program each time the step S450 is executed during the execution of the loop process when the determination result in the step S440 is negative (No). Is to do.
In this program example, by providing such step S450, when the determination result is negative (No), the processing in step S430 (refer to the correction table: image correction) is sequentially obtained thereafter. While it is possible to perform control to return the process to step S430 so as to be repeatedly performed on (a continuously shot image in which movement can be observed), the determination result in step S450 is affirmative (Yes). At this time, the X-ray image processing control can also be ended by assuming that the treatment by irradiation of the treatment beam for the patient 1 lying on the treatment table 62 (top plate 61) is finished.

  Hereinafter, the procedure regarding the X-ray image processing in the present embodiment will be described as (procedure: 2-1) to (procedure: 2-2) with reference to FIG.

(Procedure: 2-1)
The image acquisition (and image reacquisition when applicable) includes the following (procedure: 2-1.1) to (procedure: 2-1.2).

(Procedure: 2-1.1)
An X-ray image of the patient 1 lying on the treatment table 62 (top plate 61) is acquired (see step S410 in FIG. 17).
Specifically, (Procedure: 2-1.1) may be the same as (Procedure: 1-1) described in the first embodiment.

(Procedure: 2-1.2)
The weight (w (x)) is calculated (see step S420 in FIG. 17).
This (procedure: 2-1.2) may also be specifically the same as the contents of (procedure: 1-2) to (procedure: 1-4) described in the first embodiment. .
At the time of image reacquisition, the weight (w (x)) calculation is also reprocessed for determining the weight calculation based on the reacquired image.

(Procedure: 2-2)
Here, as image processing adaptation, the image processing of this embodiment includes the following (procedure: 2-2.1) and (procedure: 2-2.2).

(Procedure: 2-2.1)
With reference to the correction table (47A) created by the above (procedure: 2-1.1) (step S420 in FIG. 17), the image correction by the correction function as described above is executed.
This (procedure: 2-2.1) is basically the same as the content of (procedure: 1-5) described in the first embodiment.
In the image processing of the present embodiment, for the X-ray image acquired in (Procedure: 2-1.1) (Step S410 in FIG. 17) or the X-ray image acquired thereafter, (Procedure: 2-1). .2) Multiply by the weight (w (x)) obtained in step S420 in FIG.

(Procedure: 2-2.2)
It is determined whether or not there is a movement of the treatment table 62 (top plate 61) (see step S440 in FIG. 17).
Thereby, when the treatment table 62 (top plate 61) moves, it can be determined whether it corresponds when the image should be reacquired. Further, as a case to be determined here, there is a case where the patient 1 lying on the treatment table 62 (top plate 61) moves intentionally, for example, and the patient position deviates beyond a preset state. Can be targeted.
As already described, when performing trunk treatment while confirming temporal changes, including physiological changes such as breathing, heartbeat, swallowing, etc. in the patient 1 using X-ray images, such breathing, heartbeat, In addition to swallowing, when the patient 1 moves intentionally as described above and the patient position is greatly displaced, depending on the degree, it is caused by an object other than the patient 1 due to the correction function based on the image processing as described above. In some cases, reduction (such as FIGS. 13 to 15) or the like for the area Az to be performed is not effectively executed. This is also true in the case of movement control of the treatment table 62 (top plate 61) by the treatment table control device 60, due to the movement, the orientation of the image area Az or the like of the top plate in the image (on the screen) (arrangement of matrix arrangement) A similar situation can occur, for example, when the directionality with respect to the direction is inclined.
Therefore, in such a case, the X-ray image processing of this embodiment is performed so that an accurate image processing result can be obtained in accordance with the subsequent state of the treatment table 62 (top plate 61). Then, in addition to breathing, heart rate, swallowing, etc., if the patient moves intentionally and the patient's position is greatly displaced and the object image other than the patient is not reduced well, or the treatment table moves, etc. As described above, it is possible to return to the image acquisition flow in the above (procedure: 2-1).

  The image processing according to the present invention can also be performed according to the second embodiment. In the following, variations of the X-ray image processing according to the present embodiment will be described.

(Application modification of this embodiment (second embodiment))
FIG. 18 shows an applied modification of this embodiment, and is a flowchart showing an example of the contents of X-ray image processing and treatment beam irradiation control when applied to IGRT.
The portions other than the portions shown in the flowchart are the same as the corresponding step portions in FIG. 17, and therefore, a part of them is shown in FIG.

As shown in FIG. 18, in this program example, in place of step S430 in FIG. 17, a process of step S430B (image-guided radiation therapy (IGRT)) is provided, and in the preceding stage, treatment by treatment beam irradiation is performed. Step S430A is provided as a process for start (determination), and is controlled so that the process of step S430B (image-guided radiation therapy (IGRT)) can be executed on condition that the start is determined in step S430A. It is as a thing.
Thereby, as described above, only after the weight (w (x)) calculation and the correction table (47A) creation processing are performed before the treatment beam irradiation by the irradiation apparatus 30, the IGRT including the parallel processing of the following contents is performed. It is ensured that the control is started so as to be effective.

  Step S430B is parallel control consisting of X-ray image processing control (Step S430B1) and treatment beam irradiation control execution (Step S430B2), and based on continuous X-ray (fluoroscopic) imaging at a predetermined frame rate during treatment. In step S430B1, referring to the correction table, image correction is performed in real time by multiplying a sequentially acquired X-ray image by a weight (w (x)). On the other hand, in step S430B2, using the corrected image (X-ray fluoroscopic image) based on the X-ray image processing control, control is performed so that radiotherapy control by X-ray (fluoroscopic) type respiratory synchronized irradiation is performed. Become.

Therefore, the applied modification is also an example in which the image processing apparatus (method) according to the present embodiment is mounted on a radiation therapy apparatus (particle beam therapy apparatus) that performs image-guided radiotherapy (IGRT) (to the IGRT apparatus). It is also an example of application).
This embodiment may be implemented in this way.
In addition, also about this embodiment (2nd Embodiment) including the said application deformation | transformation form, it is also possible to implement in a form that combines one or more of each application modification form of the first embodiment. it can.

Next, an image processing apparatus (method) according to still another embodiment (third embodiment) of the present invention will be described with reference to FIGS. 19 and 20.
FIG. 19 shows an example in which processing is added to adjust the directionality when an object area other than the patient 1 is not oriented in the horizontal and vertical directions with respect to the X-ray image but is tilted. FIG. 20 is provided for explanation thereof, and shows an example in which the directionality adjustment is necessary for the direction of the top plate area Az or the like, for example.
Specifically, FIG. 20 is a diagram for explaining an example of a state where processing is necessary for adjustment related to the directionality, and the direction of the area Az such as the top plate is the image y direction (column direction). ) (Also at the same time, the image x direction (row direction)) is not parallel.

When applying image processing having a correction function according to each of the above-described embodiments (including each applied modification), an object other than the patient 1 is not necessarily aligned horizontally or vertically with respect to the X-ray image. It may not be in the state, but in some cases, each of these objects may be in different directions. In this case, by defining the coordinates with respect to the direction of each object or rotating the image itself, the image processing is performed after such a state is performed, whereby each of the above embodiments (including each applied modification) is performed. It is possible to effectively execute image processing having a correction function according to (1).
The present embodiment is based on such an idea.

For this reason, in the present embodiment, as in the example of the program shown in FIG. 19, in addition to the case of the example of the image processing program of FIG. .
Therefore, in this case, the portion composed of step S20a and step S21 (correction table creation (first processing)) corresponds to a modification of step S20 (image processing) in FIG.
In FIG. 19, step S22 (correction application (second process)) may be the same as step S22 in FIG.

In this program example, since step S20a is provided as described above, for example, as shown in FIG. 20, the direction of the area Az such as the top plate is not parallel to the column direction of the image y direction. If the image has a slight inclination, the direction of the region Az such as the inclined top plate is appropriately adjusted so as to be parallel to the y direction in the column direction by rotating the image itself. Thus, if the state shown in FIG. 5 is used, thereafter, the same image processing as in the first embodiment can be performed.
Therefore, according to the present embodiment, even if the orientation of the area Az such as the top board to be imaged due to an object other than the patient 1 is in various directions of the image matrix, the above correction is performed. It is possible to suitably perform image processing having a function, and in the same way, the image quality of the entire image can be improved, and the influence of the region caused by an object other than the patient 1 is suppressed. Image acquisition can be realized in a short time.

The image processing according to the present invention may be performed in this way.
In this embodiment (third embodiment), one or two or more of the applied variations of the first embodiment, the second embodiment, and the applied variations of the first embodiment are combined. It can be implemented in the form.
In addition, other application examples will be described below.

(Other applications: 1)
As mentioned at the beginning and as described in connection with FIG. 1 and the like, there is a patient positioning process as an application of the X-ray image.
In addition, for example, in the system 100 shown in FIG. 1, when the treatment beam B is irradiated by the irradiation device 30 to the tumor position (1a) instructed by the treatment plan, the patient position at the time of the treatment plan and the irradiation time is determined. Must be the same. If the patient position contains an error, the treatment beam B cannot be correctly irradiated to the tumor position (1a). For this reason, the patient 1 before the treatment is X-rayed by the X-ray imaging apparatus 70, and the X-ray image obtained thereby and the CT image used at the time of the treatment planning as described above (this is the reference position) are pre-treatment. A process (patient positioning) for reducing patient position error is performed.

  In this case, the image processing technique according to the present invention can also be applied to the patient positioning process performed before the treatment beam irradiation. As a result, not only the accuracy of the treatment beam irradiation timing is improved as described above. The patient positioning accuracy will be improved, and it will be effective to broaden the possibility of improving treatment results, etc., and in this respect, it will be possible to demonstrate more effectiveness in terms of radiation therapy it can.

(Other application examples: 2)
As described above, the image processing technique according to the present invention is based on, for example, an algorithm for improving the image quality of a fluoroscopic image in real time, and is mounted on a radiotherapy apparatus using the fluoroscopic image, thereby improving the treatment accuracy. It is what makes it possible to improve.
Thus, the radiotherapy has been mainly described as described above. However, the present invention is not limited to this, and for example, diagnosis can be performed by adapting to a diagnostic region such as angiography that performs X-ray image acquisition from various angles. It is also possible to improve performance.

1 ... Patient (subject, subject to be irradiated)
DESCRIPTION OF SYMBOLS 40 ... Control apparatus 42 ... Image processing control part 47 ... Memory | storage part 47A ... Correction table 70 ... X-ray imaging apparatus (X-ray imaging system)

Claims (9)

An apparatus for image processing having a correction function for improving the image quality of an image obtained by X-ray imaging of a subject,
One or more of the subject, a treatment table on which the subject lies, a cover of an irradiation device that irradiates the subject with a treatment beam, or a rotating bed of a gantry treatment room in which the treatment table is disposed Means for continuously acquiring images obtained by X-ray imaging of an object other than a subject;
Based on the image obtained by the means, a group of pixels extending along the arrangement in the image matrix array is used as a unit of processing, and for each processing unit, common to all pixel values to be a target belonging to the processing unit. A means for calculating and determining a weight to be multiplied, and as an index for determining the weight, the standard deviation of the entire image is minimized, or any one of variance, average value, entropy, and root mean square is used. Means for calculating and determining weights, and repeatedly executing the calculation determination processing for each processing unit of the image matrix array sequentially while changing the position of the processing unit. When,
Means for storing a weight for each processing unit determined by the means in a correction table in association with information on a position of each corresponding processing unit;
Means for obtaining a corrected image by applying the weight stored in the correction table by the means to the image to be corrected;
The means for obtaining the corrected image applies the weight as a correction multiplication coefficient to an image sequentially newly acquired at a predetermined frame rate for the subject after the weight is stored in a correction table. An image processing apparatus.   In the image correction by the means for obtaining the corrected image, it is checked whether or not the image reacquisition by the means for obtaining the image is necessary, and if it is determined to be necessary, the image is reacquired and the reacquired image is obtained. The image processing apparatus according to claim 1, further comprising means for newly performing a weight calculation determination process based on the process.   The image reacquisition is performed when the treatment table moves or when the position of the subject on the treatment table deviates beyond a preset state. Image processing device.   As a result of X-ray imaging of the subject, if the region imaged due to an object other than the subject is not oriented in the direction along the array in the image matrix array, the region is aligned in the array direction of the image matrix array. The image processing apparatus according to claim 1, further comprising a unit that adjusts the directionality along the direction.   5. The image processing according to claim 4, wherein the means for adjusting the directionality adjusts the directionality by rotating a pre-correction image or defining coordinates for the object. apparatus.   In the calculation determination process in which the weight is determined by calculating and determining a weight for each processing unit for each processing unit, the start position of the processing is determined from an image end or an arbitrary processing unit. 6. The image processing apparatus according to claim 1, wherein the weight calculation determination process is executed starting from the position and continuously changing the position to adjacent processing units.   The image processing apparatus according to claim 1, wherein in the means for determining the weight, the direction of the array to be the processing unit is a body axis direction of the subject. An image processing apparatus according to any one of claims 1 to 7,
Radiotherapy beam irradiation means for irradiating a patient as a subject of radiotherapy with a radiotherapy beam,
Means for determining the start of treatment by irradiation with the radiation treatment beam;
On the condition of the start determination, image processing in the image processing apparatus and radiotherapy beam irradiation control in the radiotherapy beam irradiation means are executed, and the subject is obtained by X-ray imaging at a predetermined frame rate during radiotherapy. The image processing apparatus sequentially acquires images to be corrected based on the continuous X-ray fluoroscopy, the correction table is referred to by the image processing apparatus, and the images acquired sequentially by the continuous X-ray fluoroscopy Radiotherapy control by X-ray fluoroscopic respiration synchronized irradiation using the radiotherapy beam irradiation means using a corrected X-ray fluoroscopic image based on image processing by the image processing apparatus in real time Means for controlling the image processing device and the radiation therapy beam irradiation means to perform:
A radiotherapy apparatus comprising:   A program for causing a computer to implement the image processing apparatus according to any one of claims 1 to 7 or the radiotherapy apparatus according to claim 8.