WO2015072072A1

WO2015072072A1 - Image processing device and method for operation thereof, imaging device and method for operation thereof, imaging system, and computer program

Info

Publication number: WO2015072072A1
Application number: PCT/JP2014/005145
Authority: WO
Inventors: 辻井　修
Original assignee: キヤノン株式会社
Priority date: 2013-11-18
Filing date: 2014-10-09
Publication date: 2015-05-21
Also published as: JP6344902B2; JP2015097590A

Abstract

Provided is a further method for detecting neoplasia such as cancer from fluoroscopic images of breasts. A plurality of X-ray images are obtained which include a first X-ray image that is obtained by imaging a breast at a first timing and a second X-ray image that is obtained by imaging the breast at a second timing where blood flow is less than the first timing. An image is generated on the basis of difference in regions that include mammary glands from the first X-ray image and the second X-ray image.

Description

Image processing apparatus and operating method thereof, imaging apparatus and operating method thereof, imaging system, and computer program

The present invention relates to an image processing apparatus and its operating method, an imaging apparatus and its operating method, an imaging system, and a computer program, and more particularly to a technique for detecting a neoplasm such as cancer from the breast.

As a method for detecting a neoplasm such as cancer from the breast, a radiographic method is widely used, and a mammography method is generally used. In recent years, in mammography photography, a method of taking a moving image while moving the X-ray generator (Patent Document 1) and a method of creating a tomosynthesis image by taking X-ray photography from a plurality of directions (Patent Document 2) have been further developed. ing.

Meanwhile, a technique for detecting pulmonary embolism using a plurality of X-ray images is also known. In the X-ray image, the pixel value of the blood vessel portion varies with changes in blood flow in accordance with the heartbeat. Further, in a blood vessel in which an embolus exists, a change in blood flow volume corresponding to the heartbeat is small. In Patent Document 3, the change of the pixel value in the X-ray image of the lung is presented to the user, thereby enabling the user to determine the embolic site.

JP 2004-166834 A Special table 2012-512669 gazette International Publication No. 2007/0778012

The mammography method has a problem that a calcification or a tumor, that is, cancer, is not visible in an X-ray image particularly when the breast as a subject has a large number of mammary glands (dense breast).

The present invention provides a further method for detecting a neoplasm such as cancer from a fluoroscopic image of the breast.

In order to achieve the object of the present invention, an image processing apparatus according to an embodiment of the present invention comprises the following arrangement. That is,
A first X-ray image obtained by imaging a breast at a first timing, and a second X-ray image obtained by imaging the breast at a second timing with less blood flow than the first timing Obtaining means for obtaining a plurality of X-ray images including:
Generating means for generating an image based on a difference between regions including mammary glands from the first X-ray image and the second X-ray image;
It is characterized by providing.

A further method for detecting a neoplasm such as cancer from a fluoroscopic image of the breast can be provided.

Other features and advantages of the present invention will become apparent from the following description with reference to the accompanying drawings. In the accompanying drawings, the same or similar components are denoted by the same reference numerals.

The accompanying drawings are included in the specification, constitute a part thereof, show an embodiment of the present invention, and are used to explain the principle of the present invention together with the description.
1 is a block diagram illustrating an example of an imaging apparatus according to Embodiment 1. FIG. FIG. 5 is a diagram illustrating an example of a process according to the first embodiment. FIG. 4 is a block diagram illustrating an example of an imaging apparatus according to

Embodiments

2 and 3. FIG. 10 is a diagram illustrating an example of a process according to the second embodiment. FIG. 10 is a diagram illustrating an example of a process according to a third embodiment. 5 is a flowchart of an example of processing in the first embodiment.

The present inventor has found that a change in blood flow according to the heartbeat occurs at a site where a neoplasm such as cancer exists in the breast, and that this change in blood flow can be detected by radiography. It came to complete. Embodiments of the present invention will be described below with reference to the drawings. However, the scope of the present invention is not limited to the following examples.

[Embodiment 1]
FIG. 1 is a block diagram of the photographing apparatus according to the first embodiment. The imaging apparatus according to Embodiment 1 performs X-ray imaging of a subject, and includes an X-ray generation unit 1, an X-ray detection unit 3, a compression unit 4, a pulsation detection unit 5, and a calculation unit 6. In addition, the photographing apparatus is connected to the display unit 7. The X-ray generator 1 generates a plurality of X-ray pulses. The generated X-ray pulse passes through the subject and reaches the X-ray detector 3. The X-ray detection unit 3 images the X-rays that have arrived for each X-ray pulse, and transfers the obtained image to the calculation unit 6. That is, X-ray imaging is performed using the X-ray generator 1 and the X-ray detector 3. Hereinafter, an operation method of such an imaging apparatus will be described.

The imaging device including the X-ray generation unit 1, the X-ray detection unit 3, the compression unit 4 and the pulsation detection unit 5 and the image processing device including the calculation unit 6 may be separate devices. In this case, the photographing apparatus and the image processing apparatus constitute a photographing system.

Although the subject is not limited to a specific organism, it will be described below as a human. In the present embodiment, the photographing apparatus performs mammography photographing. That is, X-ray imaging is performed while the compression unit 4 is compressing the breast 2 of the subject. In the present embodiment, an X-ray fluoroscopic image of the breast is captured by a mammography method, but the imaging method is not limited to this method. For example, a fluoroscopic image of the breast can be taken using any suitable radiation. In this specification, a fluoroscopic image refers to an image captured using radiation or the like that passes through a subject, and the fluoroscopic image includes a normal X-ray image captured using an X-ray imaging apparatus. . Further, photographing may be performed without using the mammography method, that is, without compressing the breast. The subject is not limited to the breast. For example, since there are many new blood vessels in a cancer site, not limited to breast cancer, a fluoroscopic image of a site other than the breast may be taken to detect a neoplasm.

The pulsation detecting unit 5 detects the pulsation of the subject. In the present embodiment, the pulsation detection unit 5 is connected to the compression unit 4. The pulsation detection unit 5 detects a pulsation in the breast 2 compressed by the compression unit 4, that is, a pulsation of a blood vessel included in the breast 2. The pulsation detection unit 5 may detect the pulsation of the subject in the vicinity of the breast 2 or in a part other than the breast 2 or the vicinity of the breast 2. However, the pulsation detection unit 5 can more accurately synchronize the pulsation of the blood vessels of the breast 2 and the timing of X-ray irradiation by detecting the pulsation in the breast 2 or in the vicinity of the breast 2. The pulsation detection unit 5 can detect pulsation using, for example, an oxygen saturation meter using near infrared rays. However, the pulsation detection method by the pulsation detection unit 5 is not particularly limited. For example, the pulsation detection unit 5 can detect the pulsation of the subject using an electrocardiograph (not shown), or can detect the pulsation using a vibration sensor in contact with the breast 2. Thus, since the pulsation synchronizes with the change in the blood flow, the state of the blood flow can be specified by detecting the pulsation.

The X-ray generator 1 generates at least two pulses of X-rays. In the present embodiment, each X-ray pulse is exposed in synchronization with the pulsation detected by the pulsation detection unit 5. Specifically, the pulsation detection unit 5 determines a first timing and a second timing with a smaller blood flow volume than the first timing from the detected pulsation. Specifically, a timing at which the blood flow volume in a predetermined region within the imaging region of the subject imaged by the X-ray detection unit is less than the first timing is determined as the second timing. For example, in an embodiment in which imaging of the breast 2 such as mammography imaging is performed, a timing at which the blood flow rate in a predetermined region of the breast 2 is less than the first timing is determined as the second timing. Assuming that the blood flow in each region of the breast 2 changes in the same manner, it can be said that the blood flow in the breast 2 is smaller than that in the first timing at the second timing.

In order to reduce noise in the difference image, in one embodiment, a timing with a larger blood flow is determined as a first timing, and a timing with a smaller blood flow is determined as a second timing. For example, the first timing may correspond to the systole and the second timing may correspond to the diastole. In the following, the first timing with more blood flow than the second timing is referred to as “timing with high blood flow”, and the second timing with less blood flow than the first timing is referred to as “timing with low blood flow”. .

Then, the pulsation detection unit 5 transmits a timing signal indicating these timings to the X-ray generation unit 1. In this way, the X-ray generator 1 emits X-rays at least at a timing when the blood flow in the breast 2 is high and at a timing when the blood flow is low.

In this way, a first fluoroscopic image (hereinafter referred to as a systolic image) of the breast 2 is obtained by photographing the breast 2 when the blood flow is large, and the breast 2 is photographed by photographing the breast 2 when the blood flow is small. The second fluoroscopic image (hereinafter referred to as a diastole image) is obtained. As will be described later, in order to compare the systolic image and the diastolic image, a difference image between the systolic image and the diastolic image is generated. For this reason, the systolic image and the diastole image are taken from the same predetermined direction.

The acquisition unit (not shown) included in the calculation unit 6 includes the first fluoroscopic image (systolic image) obtained by imaging the breast at the first timing and the second timing obtained as described above. A plurality of fluoroscopic images including a second fluoroscopic image (diastolic image) obtained by photographing the breast are obtained. And the production | generation part (not shown) with which the calculating part 6 is provided produces | generates the image based on the difference of a systole image and a diastole image. In the embodiment in which the breast 2 is imaged, the image based on this difference includes a region including the mammary gland. In the present embodiment, the generation unit generates a difference image obtained by performing a subtraction process between a systolic image and a diastole image as an image based on the difference. However, in another embodiment, the generation unit may generate an image obtained by further performing arbitrary image processing on the difference image as an image based on the difference. The difference image obtained in this way may be a small signal in addition to the location of blood vessels that normally exist in the breast 2, but if present, it indicates the location of new blood vessels. The presence of such new blood vessels suggests the presence of breast cancer (for example, microcalcification or mass).

A display control unit (not shown) included in the calculation unit 6 causes the display unit 7 to display the obtained difference image. The user can find the location of breast cancer by referring to the difference image. In particular, by combining a conventional X-ray image interpretation technique and a difference image, it becomes possible to detect breast cancer with higher accuracy than before. For example, the calculation unit 6 may perform processing such as specifying the presence of breast cancer or the location thereof by performing further processing on the difference image. In this case, the display control unit can cause the display unit 7 to display a result image including the processing result.

Image processing performed by the calculation unit 6 will be described along the flow shown in FIG. In FIG. 2, the case where the pulsation detection unit 5 acquires an electrocardiogram 8 obtained by an electrocardiograph and detects the timing of pulsation in the breast 2 according to the electrocardiogram 8 will be described. From the electrocardiogram 8 obtained by the electrocardiograph, the systole and diastole of the heart can be detected. Specifically, the period when the potential is large is the systole, and this corresponds to the timing when the blood flow (that is, the blood volume) flowing through the blood vessel is large as shown in the perspective image 10 in the systole. On the other hand, the timing at which the potential is flat is the diastole, and as shown in the perspective image 9 in the diastole, this corresponds to the timing when the blood flow (that is, the blood volume) flowing through the blood vessel is small. Thus, the state of the blood flow rate can be specified by referring to the electrocardiogram 8.

Electrocardiogram 8 shows the timing when blood flow is sent from the heart. Therefore, in consideration of the time required for blood flowing in the blood vessel (particularly an artery) to reach the breast 2 from the heart, the timing for exposing two X-ray pulses is determined. Then, an imaging control unit (not shown) included in the arithmetic unit 6 captures a systolic image when the blood flow rate of the breast 2 is large, and so as to capture a diastolic image when the blood flow rate of the breast 2 is small. The line generator 1 is controlled. The time required for blood to reach the breast 2 from the heart depends on the age and physique of the subject of the subject. This time can be determined empirically based on patient data. Here, in order to simplify the description, it is assumed that this time is zero.

In the systolic image, since the blood flow in the blood vessel (particularly the artery) is increased as compared with the diastole image, the transmitted dose of X-rays in the blood vessel portion is reduced. When comparing the timing when the blood flow flowing through the blood vessel is large and the timing when the blood flow flowing through the blood vessel is small, the increase in blood flow in the neovascular portion is particularly noticeable. Therefore, by creating the difference image 11 between the systolic image and the diastolic image, an image suggesting the presence of breast cancer can be created. Specifically, in the systolic image, blood flow in blood vessels (particularly arteries) increases, so that the amount of X-ray transmission in a breast cancer portion (new blood vessel portion) is relatively small. When the pixel value of the image is larger as the X-ray transmission amount is larger, the pixel value of the breast cancer portion 23 (neovascular portion) becomes negative when the pixel value of the diastolic image is subtracted from the pixel value of the systolic image.

However, considering that the difference between the pixel values of the systolic image and the diastolic image is very small, in order to perform detection more accurately, the pixel values of the two images are set in accordance with the X-ray radiation dose. It can be corrected. In general, there is a deviation of several percent in the radiation dose of each X-ray pulse emitted from the X-ray generator 1. In order to compensate for this shift, in one embodiment, the calculation unit 6 includes a radiation dose detection unit (not shown) that detects the radiation dose emitted when the systolic image and the diastolic image are captured. .

For example, by using a reference detector (not shown) attached in the vicinity of the X-ray generator 1 as a measuring device, the actual radiation dose of each pulse can be measured. Such a reference detector is arranged at a position where an X-ray pulse irradiated when a systolic image and a diastole image are captured. The radiation dose detection unit can acquire the radiation dose irradiated when capturing the systolic image and the diastolic image from such a reference detector.

As another method, the actual radiation dose can be detected based on the pixel value in the correction data area. The correction data area is an image area corresponding to an X-ray that does not transmit through the subject as shown in FIG. 2, and is an area in which the breast 2 is not shown in the present embodiment. In the example of FIG. 1, the correction data region is a region of an image taken in the region opposite to the subject's chest wall in the X-ray detection unit 3. In one embodiment, the radiation dose detection unit detects the irradiated radiation dose or a change in the radiation dose based on the average pixel value in the correction data area.

Based on the X-ray radiation dose detected in this way, the correction unit (not shown) included in the calculation unit 6 compensates for the difference in the radiation dose irradiated when the systolic image and the diastolic image are captured. As described above, at least one pixel value of the systolic image and the diastolic image is corrected. Thus, it is possible to estimate a systolic image and a diastolic image obtained when an X-ray pulse having the same radiation dose is exposed. As a specific correction method, the radiation dose of the X-ray pulse exposed to capture the diastolic image is a times the radiation dose of the X-ray pulse exposed to capture the systolic image. In some cases, there is a method of dividing the pixel value of the diastole image by a.

The difference image is an image of an area including the mammary gland. Since breast cancer occurs in the mammary gland portion, it can be detected from an image of an area including the mammary gland. Although the difference image may be an image of the entire breast including the mammary gland, in one embodiment, the difference image is an image of an area including the mammary gland portion selected from the entire breast. While the difference image includes a noise component, the pixel value indicating the breast cancer portion in the difference image is very small. Therefore, the noise portion can be reduced by extracting a region including the mammary gland portion from the difference image.

Since the mammary gland portion has a small amount of X-ray transmission, a region having a radiation transmission amount smaller than a predetermined threshold can be extracted as a mammary gland region. Specifically, the region of the mammary gland portion can be extracted from the difference image by binarizing the X-ray image using an appropriate threshold value. The determination of the mammary gland region may be performed based on either the systolic image or the diastolic image, or may be performed based on another X-ray image. Here, the image used for determining the region of the mammary gland is a fluoroscopic image of the breast 2 taken from the same direction as the systolic image and the diastolic image.

In one embodiment, the region of the mammary gland portion is determined based on the image having the larger actual radiation dose detected as described above, between the systolic image and the diastolic image. In this case, the mammary gland region 22 can be determined more accurately.

On the other hand, breast cancer occurs in the mammary gland part, but the mammary gland becomes fat with age. Therefore, in order to reduce detection omission, the mammary gland and the fat region 22 adjacent to the mammary gland may be detected as the region including the mammary gland. Specifically, the mammary gland and the mammary gland are adjacent to each other by enlarging the area of the detected mammary gland, that is, an area having a radiation transmission amount smaller than a predetermined threshold extracted from the X-ray image using a morphological operation or the like. The fat region 22 can be determined. The amount of enlargement is not particularly limited, and may be a predetermined amount determined empirically, for example. In order to erase a small signal, the reduction process may be performed before the enlargement process.

(Modification)
In the first embodiment, the X-ray detection unit 3 acquires one systolic image and one diastole image, and a difference image is generated based on the two images. In this modification, Three or more X-ray images are taken and used.

In this modification, the X-ray pulse is exposed without being synchronized with the pulsation. In this case, the X-ray generator 1 can expose a plurality of X-ray pulses at a constant cycle. In one example, the X-ray generator 1 exposes three or more X-ray pulses during one pulsation cycle. Then, the X-ray detection unit 3 generates an X-ray image for each X-ray pulse. The number of X-ray pulses exposed during one cycle can be, for example, 3 or more and 10 or less. The interval of one cycle of pulsation may be detected by the pulsation detecting unit 5, or a predetermined value (for example, 1 second) determined empirically may be used as the interval of one cycle.

In this modification, a selection unit (not shown) included in the calculation unit 6 selects a systolic image and a diastolic image from a plurality of X-ray images. The selection unit performs image processing on the obtained plurality of X-ray images, thereby obtaining an image of the breast 2 when the blood flow is increased and an image of the breast 2 when the blood flow is decreased. You can choose. For example, in the systole, since the blood flow increases, the radiation transmission amount decreases in the entire image. On the other hand, in the diastole, the blood flow decreases, and the radiation transmission amount increases in the entire image. Therefore, the selection unit selects the systolic image and the diastolic image so that the average radiation transmission amount in the diastolic image is larger than the average radiation transmission amount in the systolic image.

In one embodiment, the selection unit selects an image having the smallest amount of radiation transmission among the plurality of X-ray images as the systolic image and selects an image having the largest amount of radiation transmission as the diastolic image. Thereafter, the calculation unit 6 generates a difference image between the systolic image and the diastolic image.

When such a configuration is used, it is not essential that the photographing apparatus includes the pulsation detecting unit 5. However, such a configuration is also used when the pulsation detector 5 detects a pulsation using an electrocardiograph or when detecting a pulsation using an oxygen saturation meter attached to the fingertip. Can do. In these cases, since the timing of pulsation shifts between the pulsation detection site and the breast 2, it is difficult to accurately determine when the blood flow in the breast 2 is high and when the blood flow is low. There is a case. However, an image of the breast 2 when the blood flow is increased and an image of the breast 2 when the blood flow is decreased with reference to the period of the pulsation detected by the pulsation detection unit 5. It is possible to detect more accurately.

Also, in order to increase the S / N ratio of an image, a plurality of images with a large amount of X-ray transmission or a plurality of images with a small amount of X-ray transmission can be averaged. Furthermore, as described above, by using a reference detector, a correction data area, or the like, each X-ray image can be corrected so as to compensate for the X-ray dose variation for each X-ray pulse.

When it is difficult to increase the frame rate of the X-ray detection unit 3 to such an extent that a desired number of X-ray images can be acquired during one pulsation cycle, the following method is used. Can be used. That is, each X-ray pulse can be exposed with a different delay from the reference timing. Specifically, when it is desired to acquire an X-ray image every N times for one cycle of pulsation (the time of one cycle is T), when taking an n-th X-ray image, n is counted from the reference timing. X-ray pulses are exposed with a delay of T / N. The reference timing is a timing that is repeated at the same cycle as the cycle of pulsation, and specifically may be a systolic timing or the like. In other words, when an n-th X-ray image is taken, the X-ray pulse is exposed with a delay of mT + n · T / N (m is a natural number) from a predetermined time. According to such a configuration, even when the X-ray detection unit 3 having a small frame rate is used, N images that are substantially equivalent to N images captured during one pulsation cycle, An image can be obtained.

The calculation unit 6 may cause the display unit 7 to display an image in which the difference image is superimposed on the X-ray image of the breast 2. At this time, by displaying the difference image and the X-ray image in different colors, the user can easily detect the breast cancer site from the X-ray image. Here, the X-ray image on which the difference image is superimposed is a perspective image of the breast 2 taken from the same direction as the systolic image, the diastolic image, or the systolic image and the diastolic image.

Hereinafter, a flowchart of processing in the first embodiment or its modification will be described with reference to FIG. In step S <b> 610, the imaging control unit included in the calculation unit 6 instructs the X-ray generation unit 1 about imaging timing. In step S620, the calculation unit 6 acquires a systolic image and a diastolic image from the X-ray detection unit 3. In step S610, the imaging control unit can instruct imaging timing according to the detection result of the pulsation detecting unit 5 as in the first embodiment. On the other hand, in step S610, the imaging control unit may cause the X-ray generation unit 1 to perform imaging at a constant cycle as in the modification example. In this case, in step S620, the calculation unit 6 selects a systolic image and a diastolic image from a plurality of X-ray images.

In step S630, the calculation unit 6 generates a difference image as described above. In step S640, the display control unit included in the calculation unit 6 causes the display unit 7 to display the difference image or the result of the process performed on the difference image.

The processing by the calculation unit 6 can also be performed using a general-purpose computer. In this case, a storage medium including a computer program including instructions for causing the computer to perform processing performed by the calculation unit 6 is prepared. Then, the program included in the storage medium is loaded into a memory included in the computer, and the processor included in the computer operates according to the program, whereby the above-described processing performed by the arithmetic unit 6 can be realized.

[Embodiment 2]
FIG. 3 is a block diagram of the photographing apparatus according to the second embodiment. The imaging apparatus according to the second embodiment is the same as that of the first embodiment, but further includes an X-ray moving unit 13. Further, the processing in the second embodiment can be performed according to the flowchart shown in FIG. 6 although the acquisition method of the systolic image, the diastolic image, and the difference image is different.

The X-ray moving unit 13 moves the X-ray generating unit 1. The X-ray generator 1 exposes a plurality of X-ray pulses from multiple directions to the breast 2 while moving. The X-ray detection unit 3 captures an X-ray image in synchronization with the X-ray pulse exposure. According to this configuration, images from multiple directions can be acquired for the breast 2, and the resolution and signal quality of the image can be improved. The plurality of X-ray images thus taken include a plurality of X-ray images taken from a plurality of directions when the blood flow volume of the blood vessels included in the breast 2 is large, and a plurality of X-ray images captured when the blood flow volume of the blood vessels included in the breast 2 is small. And a plurality of X-ray images taken from the direction.

Specifically, N X-ray pulses are exposed during one cycle of pulsation, and N X-ray images are acquired. The number of X-ray pulses is not particularly limited, but may be, for example, 10 or more and 20 or less. However, the X-ray pulse may be exposed over two or more cycles of pulsation. Specific X-ray pulse irradiation control can be performed in the same manner as in the first embodiment or its modification.

The image processing performed by the calculation unit 6 will be described along the flow shown in FIG. In the present embodiment, the calculation unit 6 selects N ₁ systolic images and N ₂ diastole images from N X-ray images 14. In one embodiment, N = N ₁ + N ₂ and each X-ray image is classified as either a systolic image or a diastolic image. However, it is not necessary that all of the N X-ray images are used as the systolic image or the diastolic image. The selection of the systolic image and the diastolic image can be performed, for example, based on the detected pulsation or the magnitude of the X-ray transmission amount, as in the first embodiment or the modification thereof.

Next, a super-resolution processing unit (not shown) included in the calculation unit 6 creates a super-resolution image S1 corresponding to N ₁ systolic images. The super-resolution technique is known as a technique for creating a single image with higher resolution using a plurality of images. Also in this embodiment, super-resolution processing can be performed according to a known technique. In the super-resolution processing, in addition to the pixel information of the target image, by referring to the pixel information of the image obtained from other directions, the noise included in the target image is effectively removed, and the target image The quality of pixel information can be improved.

The super-resolution processing unit further creates a super-resolution image S2 corresponding to the N ₂ expansion period images by the same processing.

Since each systolic image or diastole image has a different shooting direction, the position of the moving subject is different in each image. In the present embodiment, the X-ray moving unit 13 moves the X-ray generating unit 1 along two axes. Therefore, in the present embodiment, super-resolution processing is performed in consideration of the biaxial movement amount of the X-ray generator 1. The moving method of the X-ray generation unit 1 is not limited to this method, and in another embodiment, the X-ray moving unit 13 moves on an arc. In this case, the super-resolution processing can be performed in consideration of the rotation angle of the X-ray generator 1.

When adopting the method of exposing an X-ray pulse delayed by n · T / N from the reference timing when taking an n-th X-ray image, which is described in the first modification of the first embodiment, the super solution This delay time needs to be taken into account when performing image processing. Specifically, when the X-ray generation unit 1 rotates and moves by an angle Θ while taking N X-ray images and the imaging interval is mT + T / N, the X-ray moving unit Rotate by / N (m is a natural number).

In the super-resolution technique, as described above, the reference image T1 is selected from N ₁ systolic images, and the reference image T2 is selected from N ₂ diastolic images. The super-resolution images S1 and S2 are obtained by improving the quality of the reference images T1 and T2 by super-resolution processing. That is, the super-resolution processing unit improves the resolution of the reference image T1 by using a plurality of fluoroscopic images of the breast 2 further photographed from a plurality of directions when the blood flow is large. In addition, the super-resolution processing unit improves the resolution of the reference image T2 by using a plurality of fluoroscopic images of the breast 2 further photographed from a plurality of directions when the blood flow is small.

As will be described later, since the difference image between the super-resolution images S1 and S2 is calculated, in one embodiment, the images obtained by exposing the X-ray pulse from the same predetermined direction are the reference images T1 and T2. Selected as. However, it is not necessary that the imaging directions (X-ray pulse exposure directions) are exactly the same, and it is only necessary that the imaging directions of the reference images T1 and T2 are substantially the same. For example, in one embodiment, the angle formed by the exposure direction of the X-ray pulse when capturing the reference image T1 and the exposure direction of the X-ray pulse when capturing the reference image T2 is equal to or less than a predetermined threshold. Thus, the reference images T1 and T2 are selected.

In addition, when the position of the X-ray generation unit 1 when the reference image T1 is captured is positioned near the center of the plurality of positions of the X-ray generation unit 1 when the N _one systolic images are captured, The quality can be greatly improved by the super-resolution processing. From this viewpoint, the reference image T1 can be selected as follows. That is, the direction of the average vector of N ₁ unit vectors representing the exposure direction of each X-ray pulse when N ₁ systolic images are captured, and the X-ray pulse when capturing the reference image T1 The angle formed by the exposure direction is set to a predetermined threshold value or less. The same applies to the reference image T2.

Thereafter, the calculation unit 6 generates a difference image using the super-resolution images S1 and S2 instead of the systolic image and the diastolic image as in the first embodiment. Similarly to the first embodiment, the mammary gland and its peripheral fat region may be extracted from the difference image.

Further, as in the third embodiment, the tomosynthesis image 17 may be reconstructed from the N X-ray images 14. The tomosynthesis image 17 can be composed of a plurality of tomographic images in a cross section orthogonal to the imaging direction of the super-resolution images S1 and S2. In this case, the calculation unit 6 can cause the display unit 7 to display an image 18 in which the difference image is superimposed on an arbitrary tomographic image selected from a plurality of tomographic images. At this time, by displaying the difference image and the tomographic image in different colors, the user can easily detect the breast cancer site from the tomographic image.

[Embodiment 3]
FIG. 3 is a block diagram of the photographing apparatus according to the third embodiment. The photographing apparatus according to the third embodiment has the same configuration as that of the second embodiment. Further, the processing in the third embodiment can be performed according to the flowchart shown in FIG. 6 although the acquisition method of the systolic image, the diastolic image, and the difference image is different.

The imaging apparatus according to the present embodiment reconstructs a tomosynthesis image by imaging the breast 2 from multiple directions. The tomosynthesis image is composed of a plurality of tomographic images of the breast 2 and is obtained by reconstructing each tomographic image from subject images taken from multiple directions. The reconstruction can be performed using a known technique that is not specifically limited. For example, a back projection method or a successive approximation method can be used.

Imaging of N X-ray images and selection of N ₁ systolic images and N ₂ diastolic images are performed in the same manner as in the second embodiment. In the present embodiment, the calculation unit 6 reconstructs a systolic tomosynthesis image C1 using N ₁ systolic images. The tomosynthesis image C1 is composed of M tomographic images C1 (m) (1 ≦ m ≦ M). The calculation unit 6 also uses the N ₂ diastolic images to reconstruct a diastolic tomosynthesis image C2 composed of M tomographic images C2 (m). The tomographic image C1 (m) in the systole and the tomographic image C2 (m) in the diastole are tomographic images on the same cross section of the breast 2. In other words, the tomosynthesis image C1 includes the first tomographic image C1 (m) in the first cross section of the breast 2, and the tomosynthesis image C2 includes the second tomographic image in the first cross section of the breast 2. It is out.

The reconstruction of the tomosynthesis image is performed in consideration of the movement amount of the X-ray generation unit 1. The amount of movement of the X-ray generation unit 1 can be considered in the same manner as the super-resolution processing in the second embodiment.

Then, the calculation unit 6 uses the systolic tomographic image C1 (m) and the diastolic tomographic image C2 (m) to generate a difference image for each cross section in the same manner as in the second embodiment. Similarly to the first embodiment, the mammary gland and its peripheral fat region may be extracted from the difference image.

In another embodiment, the calculation unit 6 may further create a tomosynthesis image of the breast 2 from N X-ray images. The calculation unit 6 may cause the display unit 7 to simultaneously display the obtained difference image or the result image obtained by further processing the difference image and the tomosynthesis image of the breast 2. In one embodiment, the computing unit 6 reconstructs a tomosynthesis image C3 composed of M tomographic images C3 (m) from N X-ray images. The tomographic image C1 (m) in the systole, the tomographic image C2 (m) in the diastole, and the tomographic image C3 (m) are tomographic images in the same cross section of the breast 2. Then, the calculation unit 6 displays on the display unit 7 an image 21 in which a difference image between the tomographic image C1 (m) in the systole and the tomographic image C2 (m) in the diastole is superimposed on the tomographic image C3 (m). Display. At this time, the difference image and the tomographic image are displayed in different colors.

(Other embodiments)
In the above-described embodiment, by injecting a contrast medium into a subject and performing X-ray imaging, a signal change due to blood can be increased, and a breast cancer portion (neovascular portion) can be detected with higher accuracy.

The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

The present invention is not limited to the above embodiment, and various changes and modifications can be made without departing from the spirit and scope of the present invention. Therefore, in order to make the scope of the present invention public, the following claims are attached.

This application claims priority based on Japanese Patent Application No. 2013-238265 filed on Nov. 18, 2013, the entire contents of which are incorporated herein by reference.

Claims

A first X-ray image obtained by imaging a breast at a first timing, and a second X-ray image obtained by imaging the breast at a second timing with less blood flow than the first timing Obtaining means for obtaining a plurality of X-ray images including:
Generating means for generating an image based on a difference between regions including mammary glands from the first X-ray image and the second X-ray image;
An image processing apparatus comprising:
The image according to claim 1, wherein the generation unit detects a mammary gland and a fat region adjacent to the mammary gland as an area including the mammary gland from an X-ray image obtained by imaging the breast. Processing equipment.
The region including the mammary gland is a region obtained by enlarging a region having a radiation transmission amount smaller than a predetermined threshold in an X-ray image obtained by imaging the breast. The image processing apparatus described.
The first X-ray image and the second X-ray image are compensated for so as to compensate for a difference in radiation dose irradiated when the first X-ray image and the second X-ray image are captured. The image processing apparatus according to claim 1, further comprising a correction unit that corrects at least one pixel value.
The correcting means is configured to determine the first X-ray image and the second X-ray based on a pixel value of a portion of the first X-ray image and the second X-ray image where a breast is not shown. The image processing apparatus according to claim 4, wherein a radiation amount irradiated when an image is taken is detected.
The correction means is arranged at a position where radiation irradiated when the first X-ray image and the second X-ray image are photographed, and the first X-ray image and the second X-ray. A radiation amount irradiated when the first X-ray image and the second X-ray image are captured is obtained from a measurement unit that measures the radiation amount irradiated when the image is captured. The image processing apparatus according to claim 4.
7. The apparatus according to claim 1, further comprising a selection unit that selects the first X-ray image and the second X-ray image from a plurality of X-ray images acquired by the acquisition unit. The image processing apparatus according to item.
The image processing apparatus according to claim 7, wherein the selection unit selects an X-ray image having an average X-ray transmission amount larger than that of the first X-ray image as the second X-ray image. .
The super-resolution processing means for improving the resolution of at least one of the first X-ray image and the second X-ray image using the plurality of X-ray images is further provided. The image processing apparatus according to any one of 1 to 8.
The image processing apparatus according to any one of claims 1 to 9, wherein the first X-ray image and the second X-ray image are obtained by mammography.
The acquisition means includes
Acquiring a plurality of X-ray images obtained by imaging the breast at a first timing, generating a tomosynthesis image including the first X-ray image from the plurality of X-ray images;
A plurality of X-ray images obtained by imaging the breast at a second timing are acquired, and a tomosynthesis image including the second X-ray image is generated from the plurality of X-ray images. Item 4. The image processing device according to any one of Items 1 to 3.
The display means further comprises display control means for displaying an image based on the difference in a color different from that of the X-ray image, superimposed on an X-ray image obtained by imaging the breast. Item 12. The image processing apparatus according to any one of Items 1 to 11.
Photographing means for photographing an X-ray image of the breast;
A specifying means for specifying the state of blood flow in the subject;
Based on the identified blood flow volume, the breast is imaged at a first timing to obtain a first X-ray image, and the breast is imaged at a second timing when the blood flow volume is lower than the first timing. Control means for controlling the imaging means so as to acquire a second X-ray image;
An imaging apparatus comprising:
14. The imaging apparatus according to claim 13, wherein the specifying means specifies a state of blood flow by detecting pulsation in a subject.
Photographing means for photographing an X-ray image of the breast;
A specifying means for specifying the state of blood flow in the subject;
Based on the identified blood flow volume, the breast is imaged at a first timing to obtain a first X-ray image, and the breast is imaged at a second timing when the blood flow volume is lower than the first timing. Control means for controlling the imaging means so as to acquire a second X-ray image;
Generating means for generating an image based on a difference between regions including mammary glands from the first X-ray image and the second X-ray image;
An imaging system comprising:
An operation method of an image processing apparatus,
A first X-ray image obtained by imaging a breast at a first timing, and a second X-ray image obtained by imaging the breast at a second timing with less blood flow than the first timing Obtaining a plurality of X-ray images including:
A generation step of generating an image based on a difference between regions including a mammary gland from the first X-ray image and the second X-ray image;
An operating method characterized by comprising:
A method of operating a photographing device,
A specific step of identifying the state of blood flow in the subject;
Based on the identified blood flow volume, the breast is imaged at a first timing to obtain a first X-ray image, and the breast is imaged at a second timing when the blood flow volume is lower than the first timing. A control process for controlling imaging timing so as to acquire a second X-ray image;
An operating method characterized by comprising:
A computer program for causing a computer to execute each step of the operation method according to claim 16 or 17.