WO2017010097A1

WO2017010097A1 - Image processing method, image processing program, image processing device, and image processing system

Info

Publication number: WO2017010097A1
Application number: PCT/JP2016/003307
Authority: WO
Inventors: 金吾七戸; 正勝帆足; 根本　茂; ▲隆▼義山口
Original assignee: 株式会社根本杏林堂
Priority date: 2015-07-14
Filing date: 2016-07-13
Publication date: 2017-01-19
Also published as: JP6670560B2; JP2017018457A

Abstract

The present invention determines a delayed contrast-enhanced CT image and a coronary CT image and accurately aligns the delayed contrast-enhanced CT image and the coronary CT image with each other, in image processing for performing difference processing using the coronary CT image and the delayed contrast-enhanced CT image, and generating a differential image. In this image processing method for: performing difference processing using a coronary CT image acquired by taking, by a CT device, an image of a left ventricle including a coronary artery with a contrast agent, and a delayed contrast-enhanced CT image acquired by taking, by the CT device, an image of the left ventricle including a myocardial infarction portion with a contrast agent after a predetermined delay time elapses since the coronary CT image has been taken; and generating a differential image, the difference processing for acquiring the coronary CT image and the delayed contrast-enhanced CT image, determining the coronary CT image and the delayed contrast-enhanced CT image, aligning the coronary CT image and the delayed contrast-enhanced CT image with each other, and subtracting the coronary CT image from the delayed contrast-enhanced CT image, is performed.

Description

Image processing method, image processing program, image processing apparatus, and image processing system

The present invention relates to an image processing method, an image processing program, an image processing apparatus, and an image processing system that perform a difference process using a coronary CT image and a delayed contrast CT image to generate a difference image.

Conventionally, it has been known that it is difficult to extract a myocardial infarction from a CT image. For example, the problem to be solved by the invention of Patent Document 1 describes that a technique for identifying a responsible blood vessel cannot extract a region where a myocardial infarction is depicted. In this Patent Document 1, in order to extract a region where a myocardial infarction is depicted from a CT image, an MR image obtained by imaging the heart with an MRI (Magnetic Resonance Imaging) device and an image obtained by a CT (Computed Tomography) device. CT images obtained are used.

JP 2010-104710 A

In myocardial infarction, fibrosis progresses from the myocardial intima side (lumen side). The contrast agent penetrates the interstitial space via the coronary artery, and the penetration rate of the contrast agent in the myocardial infarction is slower than that of the normal part of the myocardium. And the outflow rate of the contrast agent that has penetrated into the interstitial space is also slower in the myocardial infarction than in the normal myocardium. For this reason, in an image (delayed contrast image) captured in the delayed phase after a predetermined delay time has elapsed since the contrast medium arrived at the coronary artery, a difference in pixel value occurs between the normal myocardial part and the myocardial infarction part. Since this difference is large (high contrast) in the MR image, it is easier to extract the myocardial infarction compared to the CT image.

When using an MRI apparatus, imaging is performed using a load of an ATP drug (vasodilator). The MRI apparatus requires an inspection time (40 to 60 minutes) longer than the inspection time (10 to 15 minutes) of the CT apparatus. Therefore, it is desirable to identify the myocardial infarction from the CT image captured by the CT apparatus in order to reduce the burden on the patient as the subject. However, in a CT image (delayed contrast CT image) imaged in the delayed phase, the difference in pixel values between the myocardial part and the lumen is small (contrast is low), so it is difficult for a doctor to identify the boundary between the two. Furthermore, in the delayed contrast CT image, the difference in pixel values between the myocardial infarction part and the normal myocardial part is small, and it is difficult for the doctor to identify the myocardial infarction part.

Therefore, it is conceivable to perform differential processing on the CT image to emphasize the myocardial infarction portion in the delayed contrast CT image. Specifically, it is conceivable to perform a difference process in which a coronary artery CT image obtained by imaging the left ventricle including the contrasted coronary artery is subtracted from the delayed contrast CT image. Normally, the difference process is performed by subtracting an image with a low (light) average value of the left ventricle from a high (dark) image with a high average pixel value of the left ventricle to be imaged. Therefore, in general, the difference process of subtracting the dark coronary artery CT image from the light delayed contrast CT image is not performed.

By performing the above difference processing, the difference between the pixel values of the lumen of the left ventricle and the myocardium can be emphasized and the boundary between the two can be clarified. It is also possible to emphasize the difference in pixel values between the normal part of the left ventricle and the myocardial infarction. However, a new problem arises by performing this difference processing. That is, it is necessary to distinguish between a delayed contrast CT image and a coronary CT image that normally do not need to be distinguished during the difference processing. In addition, the heartbeat deforms the non-rigid heart, particularly the left ventricle. Therefore, in order to accurately identify the myocardial infarction portion of the left ventricle, it is necessary to align the delayed contrast CT image and the coronary artery CT image. However, in the delayed contrast CT image, since the boundary between the lumen of the left ventricle and the myocardial part is unclear, there is a problem that alignment with the coronary artery CT image is difficult.

In order to solve the above problems, an image processing method as an example of the present invention includes a coronary CT image obtained by imaging a left ventricle including a contrasted coronary artery with a CT device, and a predetermined amount based on the imaging of the coronary CT image. An image processing method for generating a differential image by performing differential processing using a delayed contrast CT image obtained by imaging a left ventricle including a contrasted myocardial infarction with a CT device after the delay time has elapsed. Obtaining the coronary CT image and the delayed contrast CT image, discriminating between the coronary CT image and the delayed contrast CT image, and performing alignment between the coronary CT image and the delayed contrast CT image; The difference process of subtracting the coronary CT image from the delayed contrast CT image is performed.

An image processing program as another example of the present invention includes a coronary artery CT image obtained by imaging a left ventricle including a contrasted coronary artery with a CT apparatus, and a predetermined delay time from the imaging of the coronary artery CT image. An image processing program for generating a differential image by causing a computer to execute differential processing using a delayed contrast CT image obtained by imaging a left ventricle including a contrasted myocardial infarction with a CT device after the passage And determining, in the computer, a discrimination process for discriminating between the coronary CT image and the delayed contrast CT image, an alignment process for aligning the coronary CT image and the delayed contrast CT image, and the delayed contrast CT The difference process of subtracting the coronary artery CT image from the image is executed.

An image processing apparatus as another example of the present invention includes a coronary CT image obtained by imaging a left ventricle including a contrasted coronary artery with a CT apparatus, and a predetermined delay time from the imaging of the coronary CT image. An image processing apparatus that generates a differential image by performing a differential process using a delayed contrast CT image obtained by imaging a left ventricle including a contrasted myocardial infarction portion with a CT apparatus after a lapse of time, An acquisition unit that acquires the coronary CT image and the delayed contrast CT image, a determination unit that determines the coronary CT image and the delayed contrast CT image, and alignment of the coronary CT image and the delayed contrast CT image And a difference processing unit for performing the difference process of subtracting the coronary artery CT image from the delayed contrast CT image.

As another example of the present invention, an image processing system is connected to the above-described image processing apparatus, a CT apparatus that captures the coronary CT image and the delayed contrast CT image, and injects a contrast agent. And an injection device.

Thereby, the delayed contrast CT image and the coronary CT image can be discriminated. In addition, the delayed contrast CT image and the coronary CT image can be accurately aligned, and the myocardial infarction can be accurately identified.

Further features of the present invention will become apparent from the following description of embodiments, given by way of example with reference to the accompanying drawings.

It is a block diagram of an image processing device. It is a time density curve of the coronary artery phase. It is a time density curve of a delay phase. It is the schematic explaining non-rigid body deformation alignment. It is the schematic explaining a difference process. It is a flowchart of an image process. It is the schematic of the image processing system which concerns on 2nd Embodiment. It is the schematic explaining the non-rigid body deformation position alignment which concerns on 3rd Embodiment.

Hereinafter, exemplary embodiments for carrying out the present invention will be described in detail with reference to the drawings. However, dimensions, materials, shapes, relative positions of components, and the like described in the following embodiments are arbitrary and can be changed according to the configuration of the apparatus to which the present invention is applied or various conditions. In addition, unless otherwise specified, the scope of the present invention is not limited to the embodiments specifically described below.

[First Embodiment]
An image processing apparatus 1 shown in FIG. 1 is a workstation or an image detection system, stores a CPU (control unit) that also functions as a processing unit 10 that processes an image, an image processing program, and stores image data. And a storage unit 11. The storage unit 11 includes a RAM that is a system work memory for the processing unit 10 to operate, a ROM that stores an image processing program, a control program, system software, and the like, a hard disk drive, and the like.

The CPU controls the entire image processing apparatus 1 based on the program stored in the storage unit 11. That is, the CPU executes various processing operations such as calculation, control, and discrimination according to the stored control program. Note that the CPU can also execute processing operations according to a program stored in an external storage medium such as a CD (Compact Disc) or a server on the Internet.

The image processing apparatus 1 includes a keyboard for inputting predetermined commands or data, or an input unit 12 including various switches, and a display unit for displaying a CT image, an input state of the apparatus, a setting state, a measurement result, various information, and the like. 13. Note that the image processing apparatus 1 may include a touch panel that functions as both the input unit 12 and the display unit 13.

The image processing apparatus 1 is a computer that executes various processes according to the image processing program according to the first embodiment. Therefore, in the processing unit 10, each unit such as the acquisition unit 14, the determination unit 15, the extraction unit 16, the calculation unit 17, the alignment unit 18, and the difference processing unit 19 is logically realized as various functions. The image processing program causes the image processing apparatus 1 to function as the acquisition unit 14, the determination unit 15, the extraction unit 16, the calculation unit 17, the alignment unit 18, and the difference processing unit 19. This image processing program can be recorded on a computer-readable internal or external recording medium.

Further, the image processing apparatus 1 is wirelessly or wired connected to a PACS (Picture Archiving and Communication Systems) 4 as an external storage device (server). The image processing apparatus 1 can transmit information to the PACS 4 and can receive information from the PACS 4. The image processing apparatus 1 may be connected to another external storage device such as RIS (Radiology Information System) or HIS (Hospital Information System). Further, the image processing apparatus 1 is wirelessly or wired connected to the CT apparatus 3, and the CT apparatus 3 is wirelessly or wired connected to the contrast medium injection apparatus 2. The CT apparatus 3 and the injection apparatus 2 are connected to the RIS 5 by wireless or wired connection.

The injection device 2 generates an injection protocol and injects a contrast medium and physiological saline into the body of a patient who is a subject according to the injection protocol. The CT apparatus 3 captures the left ventricle including the contrasted coronary artery to obtain a coronary artery CT image. The CT apparatus 3 captures the left ventricle including the contrasted myocardial infarction after a predetermined delay time has elapsed since the coronary artery CT image is captured, and obtains a delayed contrast CT image. Furthermore, the CT apparatus 3 transmits the captured CT image to the image processing apparatus 1 or the PACS 4.

Here, the imaging timing of CT images will be described with reference to FIGS. FIG. 2 shows an example of a schematic time density curve (TDC) including when the coronary artery is imaged by a contrast medium (coronary artery phase). FIG. 3 shows an example of a schematic time density curve including a time when a predetermined delay time has elapsed from the coronary artery phase (delay phase). 2 and 3, the vertical axis represents the pixel value, and the horizontal axis represents the elapsed time from the start of contrast medium injection.

In FIG. 2, the imaging timing of the coronary CT image is shown by a vertical bar. This imaging timing is the timing when the coronary artery is well imaged, and in FIG. 2 is the time when 28 seconds have elapsed from the start of injection. The coronary artery is well imaged 15-30 seconds after the start of infusion. In FIG. 3, the imaging timing of the delayed contrast CT image is illustrated by a vertical bar. This imaging timing is a timing at which the difference between the pixel values of the myocardial normal portion and the myocardial infarction portion increases, and in FIG. 3, is a time point when a delay time of 300 seconds (328 seconds from the start of injection) has elapsed since the coronary artery phase. This delay time is preferably 5 to 20 minutes.

1, the acquisition unit 14 of the image processing apparatus 1 acquires a coronary CT image and a delayed contrast CT image from the CT apparatus 3 or the PACS 4. The acquisition unit 14 can also acquire the coronary artery CT image and the delayed contrast CT image stored in the storage unit 11 of the image processing apparatus 1. The determination unit 15 determines the coronary CT image and delayed contrast CT image acquired by the acquisition unit 14, and adds determination information (for example, an ID indicating the type of image) to the CT image as necessary.

The discriminating unit 15 can discriminate between a coronary CT image and a delayed contrast CT image on the basis of an average pixel value in a region (for example, a central region) including the left ventricle of the CT image or the entire CT image. Specifically, the determination unit 15 compares both CT images and determines that the one with the lower average pixel value is the delayed contrast CT image. Then, the determination unit 15 adds determination information indicating that it is a delayed contrast CT image to the delayed contrast CT image. Further, the determination unit 15 may perform determination based on the number of pixels (number of voxels) in a predetermined pixel value range in a histogram based on the pixel value and the number of pixels (number of voxels). For example, the determination unit 15 compares both CT images, and determines that the one having a smaller number of pixels (number of voxels) in the range of 200HU to 500HU is a delayed contrast CT image.

The extraction unit 16 extracts the outer contour of the left ventricle in the coronary CT image (hereinafter also simply referred to as “outer contour”) and the outer contour of the left ventricle in the delayed contrast CT image. The outer contour corresponds to the epicardial contour, and the inner contour described later corresponds to the endocardial contour. For example, the extraction unit 16 performs binarization on the CT image with an arbitrary threshold, and extracts only the heart region by extracting only the region having a large pixel value. Then, the extraction unit 16 extracts a boundary between the extracted heart region and another region as an outer contour.

The alignment unit 18 performs alignment between the coronary CT image determined by the determination unit 15 and the delayed contrast CT image. Specifically, the alignment unit 18 performs rigid body alignment between the coronary CT image and the delayed contrast CT image. For example, the alignment unit 18 specifies the center of gravity of the left ventricle of each of the coronary CT image and the delayed contrast CT image, and performs translation and rotation of the coronary CT image so that the centers of gravity coincide. The alignment unit 18 may perform parallel movement and rotational movement of the delayed contrast CT image.

In addition, when aligning, the alignment unit 18 may perform rigid body alignment so that the positional deviation between the outer contours of the left ventricle of the coronary CT image and the delayed contrast CT image is minimized. For example, the alignment unit 18 sets a control point on the outer contour as a corresponding point among a plurality of control points set in a grid shape in both CT images. Then, the alignment unit 18 performs parallel movement and rotational movement of the coronary CT image so that the positional deviation between corresponding points on both outer contours is minimized (or within a predetermined range). Since there is a large difference in pixel values between the outer contour and the outer region, the outer contour can be accurately extracted from both the light delayed contrast CT image and the dark coronary CT image. Therefore, alignment can be performed accurately by using the outer contour that can be accurately extracted as a reference.

Further, the alignment unit 18 can also perform non-rigid deformation alignment of the coronary artery CT image and the delayed contrast CT image when performing alignment. The shape or size of the left ventricle of the coronary CT image may be different from the delayed contrast CT image due to the heartbeat or body movement. Therefore, by performing non-rigid deformation alignment, the size or shape of one image is deformed and matched with the size or shape of the other image to perform alignment more suitable for differential processing. Can do.

When performing non-rigid deformation alignment, the alignment unit 18 deforms the coronary artery CT image without deforming the delayed contrast CT image. In the delayed contrast CT image, the difference in pixel values between the normal myocardial portion and the myocardial infarction portion is larger than that in the coronary CT image, and the myocardial infarction region is imaged more accurately. Therefore, by not deforming the delayed contrast CT image, the myocardial infarction region can be accurately depicted even in the difference image obtained by the difference process. Non-rigid deformation alignment is performed, for example, by a method using a scale (similarity) based on similarity of pixel values (for example, mutual information amount or correlation coefficient). Then, the alignment unit 18 deforms the coronary artery CT image so that the degree of similarity becomes the maximum or a predetermined threshold value or more.

Non-rigid deformation alignment will be described with reference to FIG. 4 showing a schematic cross section of the left ventricle. In FIG. 4, the coronary artery CT image and the delayed contrast CT image are aligned so that the centers of gravity of the left ventricles coincide with each other. Further, the left ventricle of the coronary CT image indicated by the solid line is smaller than the left ventricle of the delayed contrast CT image indicated by the broken line. Further, the shape of the inner contour (lumen) of the coronary artery CT image is different from that of the delayed contrast CT image.

The alignment unit 18 sets a control point on the outer contour of the coronary CT image as a corresponding point C among a plurality of control points set in a grid shape in both CT images. The alignment unit 18 sets a control point on the outer contour of the delayed contrast CT image as the corresponding point D among the plurality of control points. Then, the alignment unit 18 expands the outer contour of the coronary artery CT image to the position indicated by the dotted line in FIG. 4 so that the positions of the corresponding point C and the corresponding point D substantially coincide. Thereafter, the alignment unit 18 deforms the shape of the outer contour of the coronary CT image so that the similarity between the two CT images is the maximum or equal to or greater than a predetermined threshold value. Further, the alignment unit 18 deforms the shape of the inner contour of the coronary CT image so that the similarity between the two CT images is the maximum or a predetermined threshold value or more.

On the other hand, when the left ventricle of the coronary CT image is larger than the left ventricle of the delayed contrast CT image, the alignment unit 18 sets the outer contour of the coronary CT image so that the positions of the corresponding point C and the corresponding point D are substantially coincident with each other. to shrink. Thereafter, the alignment unit 18 deforms the shape of the outer contour of the coronary CT image so that the similarity between the two CT images is the maximum or equal to or greater than a predetermined threshold value. Further, the alignment unit 18 deforms the shape of the inner contour of the coronary CT image so that the similarity between the two CT images is the maximum or a predetermined threshold value or more.

However, it is difficult to extract the inner contour of the delayed contrast CT image because the difference in pixel values between the lumen and the myocardium is small. Therefore, it is difficult to set a corresponding point in the inner contour of the delayed contrast CT image, and the inner contour of the coronary artery CT image cannot be deformed in the same manner as the outer contour. Therefore, the alignment unit 18 can change the shape of the inner contour of the coronary CT image as follows using the area or volume of the myocardial part.

That is, first, the calculation unit 17 of the image processing apparatus 1 calculates the area of the left ventricular myocardium (region shown by hatching in FIG. 4) in the coronary CT image. In the coronary artery CT image, since the difference in pixel values between the lumen and the myocardium is large, the inner contour and the outer contour can be extracted. Therefore, the calculation unit 17 can calculate the area of the myocardial part by calculating the area of the region between the extracted inner contour and outer contour. Further, the calculation unit 17 can calculate the volume of the myocardial part by performing the same process on a large number of CT images included in the volume data acquired by the acquisition unit 14. The calculation unit 17 can also calculate the ratio of the myocardial infarction portion to the myocardial portion and the width (depth) of the myocardial infarction portion from the endocardium (myocardial wall portion).

Next, the alignment unit 18 deforms the coronary CT image (inner contour of the coronary CT image) so as to maintain the calculated myocardial area or volume when performing non-rigid deformation alignment. Also in this case, the alignment unit 18 deforms the coronary CT image so that the similarity between the two CT images is the maximum or a predetermined threshold value or more. Usually, the coronary CT image and the delayed contrast CT image are imaged at the same cardiac phase using electrocardiographic synchronization. Therefore, it can be assumed that the areas and volumes of both myocardial portions are substantially the same. Therefore, by maintaining the area or volume of the myocardial part, the inner contour of the coronary CT image can be more approximated to the actual size and shape. Note that the area or volume does not need to be completely matched, and it can be considered that the area or volume is maintained even if there is an error within a predetermined range.

Difference processing using delayed contrast CT images and coronary artery CT images will be described with reference to FIG. 5 showing a schematic cross section of the left ventricle. In FIG. 5, a delayed contrast CT image is shown on the left side, a coronary artery CT image is shown in the center, and an image (difference image) after differential processing is shown on the right side. For convenience of explanation, FIG. 5 shows a CT image after alignment.

After the alignment, the difference processing unit 19 of the image processing apparatus 1 performs a difference process for subtracting the coronary artery CT image from the delayed contrast CT image. Thereby, the pixel value of the coronary artery CT image can be subtracted from the delayed contrast CT image, and the left ventricle lumen (inner contour) can be clearly depicted. That is, the pixel value of the lumen including the endocardium is subtracted, and the difference between the pixel values of the myocardial part and the lumen is emphasized. Further, the difference in pixel value between the normal myocardial part and the myocardial infarction part is emphasized, and the myocardial infarction part can be clearly depicted.

In the delayed contrast CT image, for example, the pixel value of the normal part of the myocardium is 80 HU, the pixel value of the lumen is 100 HU, and the pixel value of the myocardial infarction part is 100 HU. In the coronary artery CT image, the pixel value of the normal part of the myocardium is 50 HU, the pixel value of the lumen is 180 HU, and the pixel value of the myocardial infarction part is 20 HU. Therefore, in the difference image, the normal myocardial pixel value is 30 HU (= 80 HU-50 HU), the lumen pixel value is -80 HU (= 100 HU-180 HU), and the myocardial infarction pixel value is 80 HU ( = 100HU-20HU).

After the difference processing, the image processing apparatus 1 transmits the generated difference image to the PACS 4, and the PACS 4 stores the received difference image. Then, the stored difference image is used for interpretation by a doctor. As a result of the difference processing, there is no difference in the delayed contrast CT image, but in the difference image, the difference between the pixel values of the myocardial infarction portion and the lumen is emphasized to 160 HU. In addition, the difference in pixel value between the normal myocardial portion and the myocardial infarction portion, which was 20 HU in the delayed contrast CT image, is emphasized, and is 50 HU in the differential image. As a result, in the difference image, for example, a lumen having a relatively low pixel value is displayed in black, and a normal myocardial portion of the myocardial portion is displayed in gray. A myocardial infarction portion having a relatively high pixel value is displayed in white.

In the delayed contrast CT image, it is possible to depict the myocardial infarction and distinguish it from the normal myocardium. However, in a normal delayed contrast CT image, the lumen and myocardial infarction have close pixel values. Therefore, it is difficult to accurately identify the myocardial infarction that progresses from the endocardium. On the other hand, in the coronary artery CT image, the pixel values of only the lumen and the coronary artery are high. Therefore, in the coronary artery CT image, the lumen and the myocardial part can be easily identified. Therefore, according to the technique of subtracting the coronary artery CT image from the delayed contrast CT image, the myocardial part can be drawn easily and clearly as described above. Thereby, the doctor can identify the myocardial infarction accurately and easily.

Next, the image processing method will be described with reference to the flowchart shown in FIG. First, the acquisition unit 14 of the image processing apparatus 1 acquires a delayed contrast CT image and a coronary CT image from the CT apparatus 3 or PACS 4 (image server), and temporarily stores them in the storage unit 11 (S101). Then, the determination unit 15 determines the delayed contrast CT image and the coronary artery CT image (S102). At this time, the operator clicks the “(left ventricle) extraction button” displayed on the display unit 13. Then, the extraction unit 16 extracts the shape (range) of the outer contour and inner contour (and the myocardial portion between the outer contour and the inner contour) of the determined coronary artery CT image (S103). Then, the display unit 13 displays the extraction result on the CT image, and the operator operates the input unit 12 as necessary to correct the range of the outer contour and the inner contour. Note that the extraction unit 16 may automatically extract the shapes of the outer contour and the inner contour without depending on an instruction from the operator.

When calculating the area (volume) to perform non-rigid body deformation (YES in S104), the calculation unit 17 calculates the area (volume) of the myocardial part of the coronary CT image from the extracted myocardial part of the left ventricle. (S105). Then, the extraction unit 16 extracts the shape (range) of the outer contour of the delayed contrast CT image determined by the determination unit 15 (S106). Since the display unit 13 displays the extraction result on the CT image, the operator can correct the range of the outer contour and the inner contour.

Also, when the area is not calculated (NO in S104), the extraction unit 16 extracts the range of the outer contour of the delayed contrast CT image (S106). Since the display unit 13 displays the extraction result on the CT image, the operator can correct the range of the outer contour and the inner contour. Then, the alignment unit 18 aligns the coronary CT image and the delayed contrast CT image (S107). This alignment includes at least one of rigid body alignment and non-rigid deformation alignment.

In the case of rigid body alignment, the alignment unit 18 performs alignment so that, for example, the centers of gravity of the left ventricles specified by the outer contour match. In addition, the alignment unit 18 may perform rigid body alignment with reference to the outer contours of the left ventricle of the coronary CT image and the delayed contrast CT image so that the deviation between the outer contours is minimal or within a predetermined range. it can. The right ventricle has a large degree of deformation between the coronary CT image and the delayed contrast CT image. Therefore, more accurate alignment can be performed by performing rigid body alignment based on the outer contour of the left ventricle.

In the case of non-rigid deformation alignment, the alignment unit 18 performs alignment by deforming the outer contour of the coronary CT image so as to match the outer contour of the delayed contrast CT image, for example. At this time, the alignment unit 18 deforms the coronary CT image so that the degree of similarity is maximized. Since the delayed contrast CT image in which the myocardial infarction portion is imaged more clearly is not deformed, the myocardial infarction portion can be depicted more clearly.

Also, the alignment unit 18 can deform the coronary CT image so that the area (volume) of the myocardial part calculated in step S105 is maintained. Since all CT images are captured in the middle diastole due to the electrocardiogram synchronization, it can be assumed that the cross-sectional area (volume) of the myocardial portion in both CT images does not change significantly. Further, the outer contour of the left ventricle can be extracted relatively clearly in both the coronary CT image and the delayed contrast CT image. Therefore, at the time of non-rigid body deformation alignment, more accurate alignment can be performed by deforming the inside of the coronary artery CT image based on the area (volume) calculated from the coronary artery CT image.

After the alignment, the difference processing unit 19 performs a difference process for subtracting the coronary artery CT image from the delayed contrast CT image (S108), and the created difference image is temporarily stored in the storage unit 11. Thereby, the image processing is completed, and the display unit 13 displays the difference image. Further, the image processing apparatus 1 transmits the difference image to the PACS 4 and stores it.

In the flow, the determination unit 15 of the image processing apparatus 1 may determine a delayed contrast CT image and a coronary CT image stored in the CT apparatus 3 or the PACS 4. For example, the determination unit 15 determines a coronary CT image and a delayed contrast CT image based on the average pixel value of the CT image. When the acquisition unit 14 acquires the delayed contrast CT image and the coronary artery CT image, the discrimination information is added to both CT images and stored in the storage unit 11. During the difference processing, the delayed contrast CT image and the coronary CT image can be discriminated based on the discrimination information. In addition, the extraction process for the delayed contrast CT image and the extraction process and the calculation process for the coronary CT image can be switched in order.

This image processing is executed as follows by the image processing program. That is, after acquiring the coronary CT image and the delayed contrast CT image, the operator activates the image processing program and reads both CT images. The image processing program causes the image processing apparatus 1 to function as the determination unit 15 and executes a determination process for determining a coronary CT image and a delayed contrast CT image (S102). Next, the image processing program causes the image processing apparatus 1 to function as the extraction unit 16 and executes an extraction process for extracting the myocardial portion of the coronary artery CT image (S103). Then, the image processing program displays the extraction result on the display unit 13 of the image processing apparatus 1.

When calculating the area (YES in S104), the image processing program causes the image processing apparatus 1 to function as the calculation unit 17 and executes calculation processing for calculating the area of the myocardial portion of the coronary CT image (S105). Thereafter, the image processing program causes the image processing apparatus 1 to function as the extraction unit 16 and executes an extraction process for extracting the outer contour of the delayed contrast CT image (S106). Even when the area is not calculated (NO in S104), the image processing program similarly executes an extraction process (S106). Then, the image processing program causes the image processing apparatus 1 to function as the alignment unit 18 and executes an alignment process for aligning the coronary CT image and the delayed contrast CT image (S107).

After the alignment, the image processing program causes the image processing apparatus 1 to function as the difference processing unit 19 and executes a difference process for subtracting the coronary artery CT image from the delayed contrast CT image (S108). Then, the image processing program temporarily stores the created difference image in the storage unit 11. As a result, the image processing ends, and the image processing program causes the display unit 13 to display the difference image. Thereafter, the image processing apparatus 1 transmits the difference image to the PACS 4 for storage. The image processing program may cause the image processing apparatus 1 to execute a calculation process using a difference image so as to calculate the ratio of the myocardial infarction portion to the myocardial portion and the width of the myocardial infarction portion from the endocardium. Good.

According to the invention according to the first embodiment, it is possible to accurately discriminate between the delayed contrast CT image and the coronary artery CT image and perform the difference processing. Further, even a delayed contrast CT image having a small difference in pixel values between the lumen and the myocardial portion can be accurately aligned with the coronary artery CT image. Further, according to the invention according to the first embodiment, the boundary can be clearly depicted by emphasizing the difference in pixel values between the lumen and the myocardial part by subtraction. In addition, it is possible to clearly depict the myocardial infarction portion by emphasizing the difference in pixel values between the normal myocardial portion and the myocardial infarction portion. Thereby, the doctor can identify the myocardial infarction more accurately in the CT image having a higher resolution than the MR image. Further, the imaging by the CT apparatus 3 is shorter in imaging time than the MRI apparatus, and the burden on the patient who is the subject can be reduced when imaging the myocardial infarction. Furthermore, since the CT apparatus has a wider imaging width in one scan than the MRI apparatus, a wider range of images can be obtained by one imaging.

In the first embodiment, the example in which the coronary artery CT image and the delayed contrast CT image are two-dimensional images has been described. However, the coronary CT image and the delayed contrast CT image may be three-dimensional images (volume data). In addition, a three-dimensional difference image can be created by performing the above-described image processing on a plurality of two-dimensional images in the same manner. For example, it is not necessary to extract a myocardial infarction in all of the 320 two-dimensional images. If the threshold values for the myocardial infarction part and the normal part of the myocardium are specified for each of the coronary CT image and the delayed contrast CT image, a three-dimensional image can be created by reflecting the result in all the images. it can.

[Second Embodiment]
An image processing system according to the second embodiment will be described with reference to FIG. In the description of the second embodiment, differences from the first embodiment will be described, the same reference numerals will be given to the components described in the first embodiment, and description thereof will be omitted. Except where specifically described, the constituent elements having the same reference numerals perform substantially the same operations and functions, and the effects thereof are also substantially the same.

The image processing system 200 is connected to the image processing apparatus 1 described in the first embodiment, a CT apparatus 3 that captures a coronary CT image and a delayed contrast CT image, and a CT apparatus 3 in a wired or wireless manner and uses a contrast agent. And an injection device 2 for injection. The CT apparatus 3 includes an imaging unit 31 that images a subject according to an imaging protocol, a control device 32 that controls the entire CT apparatus 3, and a display 33. The imaging unit 31 includes a bed, an X-ray source that irradiates the subject with X-rays, an X-ray detector that detects X-rays transmitted through the subject, and the like. Then, the imaging unit 31 shoots a fluoroscopic image of the subject by irradiating the subject with X-rays and back-projecting the inside of the subject based on the X-rays transmitted through the subject. In addition, you may comprise the control apparatus 32 and the display 33 integrally.

The operator reads out the examination order stored in advance by the doctor from the RIS 5 to the CT apparatus 3. This examination order includes at least one piece of information about an imaging region and an imaging method (imaging timing such as a coronary artery phase and a delay phase). The test order includes information about the subject, such as the subject's weight, lean body mass, circulating blood volume, subject number (subject ID), subject name, gender, date of birth, age, height, blood volume, blood flow rate, It may include body surface area, subject disease, side effect history, creatinine value, heart rate, cardiac output, and the like. Moreover, the inspection order may include an inspection number (inspection ID), an inspection site, an inspection date, a chemical type, a chemical name, and the like as inspection information.

Then, the operator inputs the imaging region and imaging timing (coronary artery phase and delay phase) to the CT apparatus 3. Then, the CT apparatus 3 creates a coronary artery phase and a delayed phase imaging protocol. The operator confirms this imaging protocol, and if there is no correction, the imaging protocol is determined. Note that the operator may input only one of the imaging part or the imaging timing.

The imaging protocol includes information such as the imaging site, effective tube voltage, model name, manufacturer name, imaging time, tube voltage, imaging range, rotation speed, helical pitch, exposure time, dose, and imaging method. Yes. Then, the control device 32 images the subject by controlling the imaging unit 31 so as to follow the imaging protocol. The control device 32 is connected to a display 33, and the display 33 displays the input state of the device, the setting state, the imaging result, various information, and the like.

The CT apparatus 3 transmits the determined imaging protocol to the injection apparatus 2. In addition, the CT apparatus 3 can transmit a reference injection protocol to the injection apparatus 2. For example, the CT device 3 and the injection device 2 are connected via a gateway device (not shown), and the CT device 3 transmits a reference injection protocol stored in advance to the injection device 2.

The injection device 2 includes an injection head 21 in order to inject the contrast medium filled in the syringe into the body of the subject. Furthermore, the injection device 2 includes a stand 22 that holds the injection head 21 and a console 23 that is connected to the injection head 21 by wire or wirelessly. The injection head 21 has a first holding part 214 on which a syringe filled with a contrast medium is mounted, and a second holding part 215 on which a syringe filled with physiological saline for boosting the contrast medium is mounted. . In addition, the injection head 21 has a drive mechanism (not shown) that pushes out the chemical solution in the syringe mounted on the first holding unit 214 according to the injection protocol, and the chemical solution in the syringe mounted on the second holding unit 215 according to the injection protocol. And a drive mechanism (not shown) for extruding.

The console 23 functions as a control device that controls the injection head 21, and also functions as a generation device that generates a contrast medium injection protocol. Then, the console 23 generates an injection protocol based on the imaging protocol received from the CT apparatus 3. Note that the console 23 can also generate an injection protocol by modifying the reference injection protocol received from the CT apparatus 3. When the operator confirms the injection protocol displayed on the console 23, the preparation for injection is completed.

The injection device 2 may have a control device connected to the injection head 21 and a touch panel display connected to the control device and displaying the injection status of the chemical solution, instead of the console 23. In addition, the injection head 21 and the control device can be configured integrally with the stand 22. A ceiling suspension member may be provided instead of the stand 22, and the injection head 21 may be suspended from the ceiling via the ceiling suspension member.

Further, the injection device 2 may include a power source or a battery, a hand switch connected to the console 23, a remote operation device for remotely operating the injection head 21, and the like. This remote control device can start or stop injection by operating the injection head 21 remotely. Further, the power source or the battery can be provided in either the injection head 21 or the console 23, and can be provided separately.

The injection head 21 injects the contrast agent according to the injection protocol generated by the control device. This infusion protocol includes at least the infusion rate and the infusion time. The injection protocol may include information regarding injection conditions such as injection volume, injection timing, contrast agent concentration, and injection pressure. The injection head 21 has a head display 211 on which injection conditions, injection status, apparatus input status, setting status, various injection results, and the like are displayed, and an operation unit 212 for inputting the operation of the drive mechanism. is doing. The head display 211 can also be used as the operation unit 212 by being configured from a touch panel or the like. The head display 211 can be omitted.

When the chemical solution is injected, a mixing tube or the like is connected to the tip of the syringe mounted on the injection head 21. The operation unit 212 is provided with a forward button of the drive mechanism, a reverse button of the drive mechanism, a final confirmation button, or the like. When the preparation for injection such as connection of the mixing tube is completed, the operator presses the final confirmation button. Thereby, the injection head 21 stands by in a state where the injection can be started. After the start of injection, the drug solution pushed out from the syringe is injected into the body of the subject via a mixing tube or the like. This mixing tube also functions as a mixer for the contrast agent and the diluted drug solution. An example of such a mixer is “SPIRAL FLOW (registered trademark)” manufactured by Nemoto Kyorindo Co., Ltd.

The injection head 21 can be mounted with various syringes such as a prefilled syringe having a data carrier such as an RFID chip, an IC tag, and a barcode. The injection head 21 includes a reading unit that reads a data carrier attached to the syringe. The data carrier stores chemical information related to the chemical. The chemical solution information includes a product name, product ID, chemical classification, contained component, concentration, viscosity, expiration date, syringe capacity, syringe pressure, cylinder inner diameter, piston stroke, lot number, and the like.

When preparation for injection of the contrast medium is completed and the operator instructs the injection apparatus 2 to start injection, the injection apparatus 2 performs a test bolus injection of the contrast medium. The operator sets a region of interest (ROI) in advance, for example, in the ascending aorta. Then, the CT apparatus 3 images the coronary artery when, for example, 28 seconds have passed since the start of injection (coronary artery phase) using electrocardiographic synchronization. In addition, the CT apparatus 3 transmits the captured coronary artery CT image to the acquisition unit 14 of the image processing apparatus 1. Furthermore, the CT apparatus 3 uses the electrocardiogram synchronization to image the left ventricle when, for example, a delay time of 300 seconds elapses from the coronary artery phase. The CT apparatus 3 transmits the captured delayed contrast CT image to the acquisition unit 14. Test bolus injection can be omitted.

The CT apparatus 3 may transmit a coronary CT image and a delayed contrast CT image to the PACS 4. Furthermore, the CT apparatus 3 may add discrimination information to the CT image so that the discrimination unit 15 can discriminate between the coronary artery CT image and the delayed contrast CT image. Such discrimination information includes, for example, an ID indicating the type of image, a number corresponding to the order of imaging, and an imaging time.

In the image processing apparatus 1, as in the first embodiment, the determination unit 15 determines a coronary CT image and a delayed contrast CT image. The extraction unit 16 also extracts the outer contour of the left ventricle in the coronary CT image and the outer contour of the left ventricle in the delayed contrast CT image. Then, the alignment unit 18 performs alignment between the coronary CT image and the delayed contrast CT image, and the difference processing unit 19 subtracts the coronary CT image from the delayed contrast CT image. When performing non-rigid deformation alignment, the calculation unit 17 calculates the area or volume of the myocardium in the left ventricle of the coronary CT image.

Also according to the invention according to the second embodiment, it is possible to accurately discriminate between the delayed contrast CT image and the coronary artery CT image and perform the difference processing. Further, even a delayed contrast CT image having a small difference in pixel values between the lumen and the myocardial portion can be accurately aligned with the coronary artery CT image. In addition, the boundary can be clearly identified by emphasizing the difference in pixel values between the lumen and the myocardium by subtraction. In addition, it is possible to clearly depict the myocardial infarction portion by emphasizing the difference in pixel values between the normal myocardial portion and the myocardial infarction portion.

[Third Embodiment]
Image processing according to the third embodiment will be described with reference to FIGS. 4 and 8. In the third embodiment, when performing non-rigid deformation alignment, the inner contour of the left ventricle in the coronary CT image is deformed according to the amount of displacement of the outer contour of the left ventricle in the coronary CT image.

In the description of the third embodiment, differences from the first embodiment and the second embodiment will be described, and the same reference numerals will be assigned to the components described in the first embodiment and the second embodiment. The description is omitted. Except where specifically described, the constituent elements having the same reference numerals perform substantially the same operations and functions, and the effects thereof are also substantially the same.

The outer contour of the left ventricle in the coronary CT image in FIG. 4 is smaller than the delayed contrast CT image. In this case, the alignment unit 18 of the image processing apparatus 1 uses the outside of the coronary CT image so that the corresponding point C on the outer contour of the coronary CT image coincides with the corresponding point D on the outer contour of the delayed contrast CT image. Enlarge the contour. The displacement amount and the displacement direction of the outer contour accompanying this enlargement can be expressed as displacement vectors A and B as indicated by arrows in FIG.

The displacement may vary depending on the left ventricular site. For example, the displacement vector A group has a larger displacement amount than the displacement vector B group. Therefore, the alignment unit 18 deforms the inner contour of the left ventricle in the coronary artery CT image according to the amount of displacement of the outer contour. For example, the alignment unit 18 calculates displacement vectors A and B indicating to which position the corresponding point C moves. Then, the alignment unit 18 performs non-rigid deformation alignment according to the calculated displacement vectors A and B so that the similarity described in the first embodiment is the maximum or equal to or greater than a predetermined threshold. That is, the displacement vector A ′ (FIG. 8) of the portion located in the vicinity of the corresponding point C where the displacement amount is large is larger than the displacement vector B ′ of the portion located in the vicinity of the corresponding point where the displacement amount is small. Deform the inner contour.

As a result, in the inner contour of the coronary artery CT image, the portion located near the corresponding point C having a large displacement amount has a larger displacement amount than the portion located near the corresponding point C having a small displacement amount. On the other hand, when the outer contour of the left ventricle of the coronary CT image is larger than the delayed contrast CT image, the alignment unit 18 reduces the outer contour of the coronary CT image. Also in this case, the alignment unit 18 deforms the inner contour of the coronary CT image according to the displacement amount of the outer contour.

According to the invention according to the third embodiment, the inner contour can be displaced according to the displacement amount of the outer contour. Therefore, the inner contour can be more accurately deformed and the non-rigid deformation position alignment can be performed. The invention according to the third embodiment can also accurately determine the delayed contrast CT image and the coronary CT image and perform the difference process. Further, even a delayed contrast CT image having a small difference in pixel values between the lumen and the myocardial portion can be accurately aligned with the coronary artery CT image. In addition, the boundary can be clearly identified by emphasizing the difference in pixel values between the lumen and the myocardium by subtraction. In addition, it is possible to clearly depict the myocardial infarction portion by emphasizing the difference in pixel values between the normal myocardial portion and the myocardial infarction portion.

[Deformation]
When difference processing is performed using a CT image having a large difference in pixel values between a normal myocardial portion and a myocardial infarction portion, the difference in pixel values in the difference image increases. Therefore, when imaging the heart a plurality of times in one examination as in the case of capturing a perfusion image, it is preferable to use a CT image having a large difference in pixel values among a plurality of CT images. Here, the difference in the pixel values is maximized when the pixel value of the normal part of the myocardium reaches a peak. Therefore, in a modification, the acquisition unit 14 of the image processing apparatus 1 acquires a CT image captured at a time point near the pixel value peak of the normal myocardium.

For example, the pixel value of the normal part of the myocardium starts to rise 2 seconds after the pixel value of the coronary artery or the left ventricle reaches 150 HU. Then, when 7 to 8 seconds have elapsed since the arrival, the difference between the pixel values of the myocardial normal part and the myocardial infarction part reaches a peak. Thereafter, when 16 to 17 seconds have elapsed since the arrival, the difference between the pixel values reaches a level at which it is difficult to determine.

Therefore, it is preferable that the discriminating unit 15 discriminates a CT image captured under the following conditions and uses it for the difference processing. That is, with the ROI set in the ascending aorta, a CT image captured after the ROI pixel value reaches 150 HU, within a predetermined time (for example, 7 to 17 seconds) after the ROI pixel value reaches 150 HU 2 to 9 seconds after the CT image picked up at the time, the CT image picked up after the ROI pixel value started to rise (peak), and the ROI pixel value went up and started to drop It is preferable to use a CT image imaged within the range. The ROI can also be set in the coronary artery or the left ventricular lumen.

Thereby, the image processing apparatus 1 can accurately depict the myocardial infarction portion. As for the delayed phase, it is preferable to use a CT image in which the difference in pixel values between the normal myocardial portion and the myocardial infarction portion is the largest among the plurality of CT images. Further, the determination unit 15 can determine a CT image having a large difference in pixel value from a plurality of CT images based on the imaging time information added to the CT image by the CT apparatus 3. Furthermore, the determination unit 15 may calculate a difference in pixel values between a normal myocardial portion and a myocardial infarction portion of each CT image, and determine a CT image that maximizes the difference in pixel values.

As mentioned above, although this invention was demonstrated with reference to each embodiment, this invention is not limited to the said embodiment. Inventions modified within the scope not departing from the present invention and inventions equivalent to the present invention are also included in the present invention. Moreover, each above-mentioned embodiment and each deformation | transformation form can be combined suitably in the range which is not contrary to this invention.

Note that the operator can manually correct the position of the image or the shape of the outer contour after the rigid body alignment, the non-rigid deformation alignment, or the extraction of the outer contour. Further, the operator can perform image correction such as noise removal using a known method after rigid body alignment, non-rigid body deformation alignment, outer contour specification or difference processing.

Furthermore, the rigid body alignment, the non-rigid deformation alignment, and the extraction of the outer contour are not limited to the above method. For example, in non-rigid deformation alignment, corresponding points are arranged along the outer contours of the coronary CT image and the delayed contrast CT image, and the coronary CT image is deformed so as to improve the similarity between the outer contours. May be. In this case, the movement amount of the part in the coronary artery CT image other than the corresponding point can be complemented from the movement amount of the corresponding point. Further, when performing non-rigid deformation alignment, it is desirable to deform the coronary artery CT image without deforming the delayed contrast CT image, but it is sufficient that the myocardial infarction portion of the delayed contrast CT image is not deformed. Therefore, at least the inner region of the outer contour of the left ventricle of the delayed contrast CT image need not be deformed, and the outer region of the outer contour may be deformed. That is, when performing non-rigid deformation alignment, the coronary CT image can be deformed without deforming at least the inner region of the outer contour of the left ventricle in the delayed contrast CT image.

This application claims priority from Japanese Patent Application No. 2015-140289 filed on July 14, 2015, the entire contents of which are incorporated herein by reference.

1: image processing device, 2: injection device, 3: CT device, 14: acquisition unit, 15: discrimination unit, 16: extraction unit, 17: calculation unit, 18: alignment unit, 19: difference processing unit, 200: Image processing system

Claims

A coronary artery CT image obtained by imaging a contrasted left ventricle including a coronary artery with a CT device, and a left ventricle including a contrasted myocardial infarction after a predetermined delay time has elapsed since the imaging of the coronary artery CT image An image processing method for generating a difference image by performing difference processing using a delayed contrast CT image obtained by imaging with a CT apparatus,
Obtaining the coronary artery CT image and the delayed contrast CT image;
Discriminate between the coronary CT image and the delayed contrast CT image,
Alignment of the coronary CT image and the delayed contrast CT image,
An image processing method for performing the difference process of subtracting the coronary artery CT image from the delayed contrast CT image.
When performing the determination, an average pixel value of a region including the left ventricle is compared between the coronary CT image and the delayed contrast CT image, or an image between the coronary CT image and the delayed contrast CT image. The image processing method according to claim 1, wherein overall average pixel values are compared to determine that an image having a low average pixel value is the delayed contrast CT image.
The image processing method according to claim 1 or 2, wherein, when performing the alignment, non-rigid deformation alignment of the coronary artery CT image and the delayed contrast CT image is performed.
The image processing method according to claim 3, wherein the coronary CT image is deformed without deforming the delayed contrast CT image when performing the non-rigid deformation alignment.
Calculate the area or volume of the left ventricular myocardium in the coronary CT image,
The image processing method according to claim 4, wherein when performing the non-rigid deformation positioning, the coronary artery CT image is deformed so as to maintain an area or volume of the myocardial part.
6. The inner contour of the left ventricle in the coronary CT image is deformed according to a displacement amount of an outer contour of the left ventricle in the coronary CT image when performing the non-rigid deformation alignment. Image processing method.
Extracting an outer contour of the left ventricle in the coronary CT image and an outer contour of the left ventricle in the delayed contrast CT image;
The rigid body alignment is performed according to any one of claims 1 to 6, wherein, when performing the alignment, rigid body alignment is performed so that a positional shift between the outer contours of the coronary artery CT image and the delayed contrast CT image is minimized. The image processing method as described.
A coronary artery CT image obtained by imaging a contrasted left ventricle including a coronary artery with a CT device, and a left ventricle including a contrasted myocardial infarction after a predetermined delay time has elapsed since the imaging of the coronary artery CT image An image processing program for generating a differential image by causing a computer to execute differential processing using a delayed contrast CT image obtained by imaging with a CT apparatus,
Discrimination processing for discriminating between the coronary CT image and the delayed contrast CT image;
An alignment process for aligning the coronary artery CT image and the delayed contrast CT image;
An image processing program for executing the difference processing for subtracting the coronary artery CT image from the delayed contrast CT image.
A coronary artery CT image obtained by imaging a contrasted left ventricle including a coronary artery with a CT device, and a left ventricle including a contrasted myocardial infarction after a predetermined delay time has elapsed since the imaging of the coronary artery CT image An image processing apparatus that performs difference processing using a delayed contrast CT image obtained by imaging with a CT apparatus and generates a difference image,
An acquisition unit for acquiring the coronary CT image and the delayed contrast CT image;
A discriminator for discriminating between the coronary artery CT image and the delayed contrast CT image;
An alignment unit for aligning the coronary artery CT image and the delayed contrast CT image;
An image processing apparatus comprising: a difference processing unit that performs the difference process of subtracting the coronary CT image from the delayed contrast CT image.
An image processing system comprising: the image processing apparatus according to claim 9; a CT apparatus that captures the coronary CT image and the delayed contrast CT image; and an injection apparatus that is connected to the CT apparatus and injects a contrast agent. .