JP2001095795A - Medical image creation method and apparatus, and recording medium - Google Patents

Abstract

(57) [Problem] To enable more accurate creation of a three-dimensional image of a minute blood vessel lumen. In a blood vessel lumen extraction process, step S2 is performed.
At 11, the maximum pixel value of the lumen of the descending aorta is measured for each transverse image, and at step S 212, the descending aorta located in the same cross section as the target blood vessel, information on the difference between the distance from the sinus of Valsalva and the circulation time, The arrival state of the contrast agent having the maximum pixel value measured in the measurement step is corrected based on the heart rate. In step S213, a value corresponding to a predetermined ratio of a difference between a reference value and the maximum pixel value corrected in the correction step is set as a parameter serving as an extraction reference of a blood vessel lumen, and a blood vessel located in the cross section thereof is set. Used as the lower threshold for cavity extraction.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to creating a three-dimensional image of a lumen of a blood vessel from a cross-sectional image obtained by a computer tomography apparatus and the like, and more particularly to more accurate three-dimensional images of the lumen of a thin blood vessel such as a coronary artery. The present invention relates to the creation of medical images that can be created at any time.

[0002]

2. Description of the Related Art In recent years, three-dimensional images of human organs can be obtained relatively easily due to the development of various CT apparatuses, high-precision data, high-speed data acquisition, and the like. A three-dimensional image of an organ of a human body is obtained by imaging a target site using a CT apparatus, an MRI apparatus, and the like, extracting a target organ from a series of obtained cross-sectional images, and performing various image processing. It is obtained by three-dimensional reconstruction. Since the shape of the object can be intuitively recognized by the three-dimensional image created in this way, it is extremely useful for grasping the three-dimensional structure of the lesion.

For example, taking an electron beam CT apparatus as an example, pixels (pixels) forming a two-dimensional transverse image and pixels (voxels) forming a three-dimensional model have CT values (MRI).
In the case of data, it is expressed by NMR signal intensity (in the present specification, a value given to a pixel is referred to as a pixel value). Generally, CT value is -1000 Hounsfie
ld Unit (HU), water is 0 HU, and blood vessel walls and uncontrast blood are approximately 0 to 100 HU. Adipose tissue around blood vessels is approximately -100 HU. In order to obtain a three-dimensional image of the blood vessel lumen, a contrast medium is injected into the blood vessel and imaging is performed. In electron beam CT, one slice is taken for one heartbeat, so imaging time for the number of slices is required. However, the imaging state during that time differs not only between patients but also within the same patient for each cross section. Unavoidable. The contrast state depends on the concentration of the contrast agent, the timing of imaging, and the like.
As the index, the average CT value and the maximum C of the blood vessels on the same cross section
Although a T value or the like is conceivable, taking a simple maximum CT value as an example, the maximum CT value is often located at the center of the lumen (when located at the periphery, there is a high possibility of calcification and artifacts). ), Actually more than 200HU, sometimes 500
It has been found that the contrast state is not uniform because it may exceed the HU.

In the conventional three-dimensional image forming method, in order to distinguish an object from an unnecessary area when extracting a blood vessel lumen, for example, a threshold value is set, and transparency and color information (color tone) are set to each pixel value. The object is extracted in correspondence with. in this case,
Regardless of the blood vessel region and the contrast state of each cross section,
The values of the transparency or the color tone are set uniformly. For example, when extracting the coronary artery lumen, the shaded surface display (Shaded
Surface Display) method is used, but in many cases, the lower threshold of the CT value is fixed at 80 or 100 HU.

[0005]

However, as described above, the contrast state of the blood vessel lumen is not uniform, and the threshold value, transparency or color tone at the time of extracting the blood vessel lumen is fixed to a uniform value. When creating a three-dimensional image by using
Pseudo-stenosis, pseudo-defect, and creaking phenomenon in which blood vessels are elongated may occur. The present invention has been made in view of such a situation, and is intended to create a three-dimensional image of a blood vessel lumen more accurately than a conventional method.

[0006]

In order to achieve the above-mentioned object, a method of producing a medical image according to the present invention comprises the steps of obtaining a thin blood vessel from a series of cross-sectional images taken by a computer tomographic imaging apparatus or the like using a contrast agent. A medical image creating apparatus for creating a three-dimensional image of a lumen, comprising: a measuring step of measuring a maximum pixel value of an aortic lumen located in each cross section; and a parameter serving as a reference for extracting the vessel lumen. Setting a predetermined ratio of a difference between a reference value and the maximum pixel value, and extracting a blood vessel lumen. Thus, a thin blood vessel image can be accurately extracted using the value of the lumen of the aorta. An estimation step of estimating a state of arrival of a contrast agent based on information on a difference between a thin target blood vessel and the aorta, a distance between the two, and a circulation time and a heart rate from a specific site; and an arrival state obtained in the estimation step. And a correcting step of correcting the maximum pixel value measured by the measuring means, whereby a thin blood vessel image can be more accurately extracted. Thus, when a three-dimensional image is created from a series of images, the contour of the blood vessel lumen can be extracted more accurately than a conventional three-dimensional image creating method.

In the extracting step, a predetermined value of a difference between a reference value (for example, -100 HU is set as a reference value because the periphery of the coronary artery is often fat having approximately -100 HU) is used. A ratio, for example, a value corresponding to 70% can be set as the parameter.

[0008] The present invention also includes an apparatus capable of implementing the above-described method and a recording medium storing a program capable of causing a computer to implement the above-described method.

BEST MODE FOR CARRYING OUT THE INVENTION

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration example of a medical image creation device according to the present invention. In the present embodiment, it is assumed that a three-dimensional image of the coronary artery blood vessel lumen is created using the Shaded Surface Display method.

In the configuration example shown in FIG.
Reference numeral 01 denotes, for example, an electron beam CT apparatus, which images a region of interest of a subject (not shown), and obtains CT image data (hereinafter, also referred to as a transverse image, as appropriate) using a DICOM (Digita).
The image data is stored in the storage device 102 or output to the information processing device 104 in a common format of medical data called “Image and Communications in Medicine”. Storage device 102 composed of a magneto-optical disk device or the like
Temporarily stores the CT data output from the imaging apparatus 101 in the removable storage medium 103. The information processing device 104 is, for example, a workstation capable of performing three-dimensional graphics processing, and includes a memory for storing images, a CPU for controlling various peripheral devices, and the like. In the information processing apparatus 104, the image capturing apparatus 101 directly
Alternatively, CT data stored in the storage medium 103 can be input. A display device 106 (LC) for displaying an image being created, a created three-dimensional image, or the like.
D, CRT, etc.) are connected. Information processing device 104
And an input device 105 such as a keyboard for inputting instructions and the like. Thus, the information processing apparatus 104 can create a three-dimensional image based on a series of CT image data directly output from the imaging apparatus 101 or read from the storage medium 103.

Next, the operation of the medical image creating apparatus configured as described above will be described with reference to the flowchart of FIG. First, in step S111, the imaging device 101 performs imaging of a target portion of a subject. For example, a slice width of about 6 cm from the tracheal bifurcation to the lower part of the heart has a slice width of 1.
One image per heartbeat at 5 mm, for a total of 40 cross-sectional images. It is assumed that a contrast agent has been injected into the subject in advance. The obtained series of cross-sectional images are stored in the storage device 102.
Is stored in the storage medium 103. Next, step S
In step 112, the information processing apparatus 104 reads a series of cross-sectional images stored in the storage medium 103, and performs a blood vessel extraction process for each of these cross sections. Here, step S112
Will be described with reference to FIGS. 3 to 8. FIG. 3 is a diagram showing the left ventricle of the heart and the coronary artery, which is a vegetative blood vessel. In FIG. 3, for convenience of explanation,
Both atria, right ventricle, blood vessels, and the like, which are unnecessary for description, are omitted. The vertical direction in the drawing indicates the body axis direction of the subject, and the horizontal direction indicates the left and right directions of the subject. Above the left ventricle of the heart 201 (ascending aorta 202 extending cranially to the body axis)
Is connected to the descending aorta 207 via the aortic arch 206, and the sinus of Valsalva 2 which is the origin of the ascending aorta 202
From 03, right coronary artery 204 on the right and left coronary artery 205 on the left
Is growing. Ascending aorta 20 from left ventricle of heart 201
A part of the blood ejected in Step 2 flows from the sinus of Valsalva 203 to the right coronary artery 204 and the left coronary artery 205, and the rest of the blood flows to the brachiocephalic artery, the left common carotid artery, and the left subclavian artery which are omitted in the figure. As it flows, it flows to the descending aorta 207 via the aortic arch 206.

FIG. 4 shows a cross-sectional image (sectional view) taken along the line AA 'of FIG. Note that the vertical direction in the figure indicates the front (abdominal) and rear (dorsal) directions of the subject, and the horizontal direction indicates the left and right directions of the subject. In FIG. 4, the left coronary artery 205 extending from the sinus of Valsalva 203 is composed of a total of two coronary arteries, a left anterior descending branch 205A and a left circumflex branch 205B (hereinafter, the right coronary artery 204, the left anterior descending branch 205A, and the left circumflex branch 205B are simply referred to as a coronary artery). Say). When extracting a blood vessel lumen as described in the related art, a parameter (Shaded Surfa
Whether the edge of the blood vessel lumen can be accurately extracted depends on what value is set as the lower threshold of the CT value in the case of ce Display. Therefore, the present inventor measured the blood vessel diameter of the coronary artery using the following methods 1 and 2, and compared which method can obtain a more accurate blood vessel diameter. Quantitative coronary angiography (QCA) using existing coronary angiography data performed the day before electron beam imaging as reference data for coronary artery diameter
raphy) was used.

The blood vessel diameter at point C in FIG. 4 is measured. From a series of cross-sections, a multi-section reconstruction method (MPR: Multiplana) for calculating and displaying a cross section in a direction different from the original cross section
r Reconstruction) is used to create a plane corresponding to the long axis plane of the blood vessel at point C in FIG. 4 (see FIG. 5). In the method 1, a short axis plane (shown in FIG. 6) perpendicular to the long axis of the blood vessel at the point C in FIG. The diameter a in the same reading direction as that of the QCA in the cross section was defined as the blood vessel diameter by the electron beam CT. Window wid
Set th to 0 and window level to 80, 100, 12 respectively
By setting 0,140 HU, it corresponds to the blood vessel diameter when the lower limit threshold (hereinafter simply referred to as a threshold) of the CT value at the time of extracting the blood vessel lumen in the Shaded Surface Display is set to 80, 100, 120, 140 HU, respectively. Thought. When the average and standard deviation of the difference between the measurement result and the corresponding reference data (QCA data) were calculated under four conditions, the average value of the difference was closest to 0 under the condition corresponding to the case where the threshold was set to 100 HU. The result was obtained. The standard deviation showed no difference between the four conditions.

On the other hand, in method 2, a profile curve showing the distribution of CT values in the direction perpendicular to the long axis of each part is obtained. For example, FIG. 7 shows a long axis plane of the blood vessel at the point C, and FIG. 7 shows a profile curve showing a distribution of CT values on a line BB ′ perpendicular to the long axis of the blood vessel (as a result, QC
A is the same as the reading direction). Using the value of the substance around the object as a reference value, the distance between the curves having a height corresponding to 50, 60, 70, and 80% of the difference between the left and right reference values and the maximum CT value in the profile curve (see FIG. In 7, the reference value is -100H
The distance b between the curves having a height corresponding to 70% of the difference between U and the maximum CT value is shown), and the average of the difference between the measured value and the reference data (QCA data) is measured in the same manner as in Method 1. And the standard deviation were calculated. The result obtained was that the distance between the curves having a height corresponding to 70% of the maximum CT value was the closest to zero. The standard deviation showed no difference between the four conditions.

The standard deviation of the difference in the method 2 is significantly smaller than that in the method 1, so that the edge of the blood vessel lumen can be correctly extracted in order to uniformly extract the edge of the blood vessel. Rather than setting a proper threshold, the reference value and the maximum C
It has been found that a method of setting a value of a predetermined ratio of the difference from the T value (here, 70% is optimal) as a lower limit threshold at the time of blood vessel extraction, that is, a method of using a site-specific threshold is more appropriate.

However, it is impossible to set a threshold value for each coronary artery of each patient in the actual clinical setting using the above-described profile curve method for each coronary artery of each patient. Therefore, referring to FIG. 4 again, the descending aorta 207 always exists in the transverse image in which the coronary artery exists.
It was thought that it would be easier to measure this maximum CT value and substitute it for the maximum CT value of the coronary artery present on the same cross section. However, in the contrast state of the coronary artery and the descending aorta 207, as shown in FIG. 3, there is a difference in the distance between the two from the sinus of Valsalva, and there is a slight time phase shift for the arrival of the contrast agent. for that reason,
By correcting the difference between the distance from the sinus of Valsalva, the circulation time, and the heart rate, almost the same time phase can be obtained. That is, for each transverse image, the maximum CT value of the lumen of the descending aorta (generally located at the center of the lumen) on the same transverse image is measured, and the approximate difference in distance, circulation time, heart rate, etc. The state of arrival of the contrast agent is corrected based on the information of the coronary artery, and the corrected value is regarded as the maximum CT value of the coronary artery on the same cross section, that is, an index indicating the contrast state of the same cross section, and the reference value and the corrected maximum value A predetermined ratio of the difference from the CT value, for example, a value corresponding to 70% can be adopted as the lower threshold of the coronary artery existing on the cross section.

Based on the above, the processing shown in FIG. 8 is executed in the blood vessel lumen extraction processing in step S112. Step S
In 211, the information processing apparatus 104 measures the maximum CT value located substantially at the center of the lumen of the descending aorta for each transverse image. At this time, the user (examiner) specifies the center of the lumen of the descending aorta (a part that is clearly not calcified or an artifact) for each transverse section using the input device 105 or another pointing device, and performs information processing. The device 104 sets the CT value of the designated portion as the maximum CT value. Set the lower threshold to 80HU, 100HU, etc.
The display creates an approximate three-dimensional image of the descending aorta, the user manually specifies the starting point and the end point of the descending aorta in the image, and the information processing apparatus 104 automatically calculates the maximum CT value for each transverse section of the descending aorta. It is also possible to make a measurement. Then, the information processing device 104 corrects the arrival condition of the contrast agent having the maximum CT value measured based on the difference in the distance between the descending aorta and the coronary artery from the sinus of Valsalva on the same cross section and the information such as the circulation time and the heart rate. (Step S212). Subsequently, a predetermined ratio of the difference between the reference value specified in step S213 (here, specified as -100HU) and the corrected maximum CT value, for example, a value corresponding to 70% is set as a threshold value of the coronary artery located on the same cross section. And extract the lumen of the blood vessel.
FIG. 9 shows an example of the threshold value of the coronary artery set in the transverse images of transverse section numbers 1 to 40 at this time. In FIG. 9, a polygonal line P indicates the maximum CT value of the descending aorta included in each cross-sectional image. As can be seen in FIG. 3, the coronary artery and the descending aorta of the same cross section are much closer to the sinus of Valsalva. The blood circulation time depends on the cardiac output, etc., but is normal and the arm-tongue time (time from the elbow vein to the tongue artery) is 10 to 16 seconds, and the arm-lung time (the elbow vein to the lung capillaries) Is 4 to 8 seconds. In many studies, the aortic blood flow rate was normal and around 1
If the difference between the coronary artery and the descending aorta from the sinus of Valsalva is, for example, 30 cm, the contrast agent arrives at the descending aorta 0.3 second behind the coronary artery of the same cross section. Since one shot is taken per heartbeat, for example, if the heart rate is 60 beats / minute, the shot was taken every second,
That is, there is a time difference of one second between the cross sections. In this example, the maximum CT 0.3 seconds before the polygonal line P (3/10 cross-section interval before)
It can be assumed that the value approximates the maximum CT value of the coronary artery of the same cross section, and this is set as a line Q as the corrected maximum CT value. The polygonal line R indicates a value corresponding to 70% of the difference between the reference value (here, -100 HU) and the corrected maximum CT value, and this value is set as a threshold value for extracting the coronary artery lumen for each cross section. For comparison, a CT value that has been conventionally set as a threshold is shown as a straight line S. Referring to this figure, the present invention differs for each cross-sectional image (that is, according to the arrival state of the contrast agent) as compared with the conventional case in which a uniform threshold is set for all cross-sectional images. It can be seen that the threshold, that is, the threshold corresponding to each part is set.

After the extraction of the blood vessel lumen is performed in this manner, the flow returns to step S113 in FIG.
4 constructs a three-dimensional model by performing an interpolation process from an image (region extraction image) in which a blood vessel lumen is extracted from a series of cross-sectional images, and performs a projection process, a shading process, and the like on the constructed three-dimensional model. To generate a three-dimensional image. Then, the three-dimensional image (reconstructed image) created in step S114 is displayed on the display device 106.

In the above description, a case has been described in which a threshold for extracting the coronary artery lumen is set. However, similarly for other small blood vessels such as cerebral blood vessels, the maximum CT of the lumen of a thick blood vessel in the vicinity is similarly set. The threshold can be determined based on the value. In addition, for each CT value that constitutes an image, Volume and other parameters that set arbitrary transparency and color tone, etc.
The present invention can be applied to the case of the rendering technique. For example, FIG. 10 is a diagram showing the correspondence between each CT value and the transparency. FIG. 10A shows a case where an arbitrary graph is written, and FIG.
Is set to 100%, and 0HU or less and 600HU or more are set to 0%. In this example, the peak is 300 HU, but in the present invention, the value of this peak can be changed for each cross section. In this embodiment, the threshold set for CT image data obtained from an electron beam CT apparatus has been described. However, the present invention can be applied to image data obtained from an MRI apparatus or the like. it can. The computer program for performing the above-described various processes is transmitted to a user via a recording medium such as a magnetic disk, a CD-ROM, or a floppy disk, or transmitted to a user via a network or the like. Can be recorded and used.

[0020]

As described above, according to the medical image creation of the present invention, the maximum CT value of the descending aortic lumen located in the same cross section as a thin blood vessel is measured, and the maximum CT value is used as a parameter for extracting blood vessels. By setting a CT value corresponding to a predetermined ratio of a difference between a reference value and a maximum CT value for each cross section, setting a threshold value for each cross section, and extracting a blood vessel lumen, a three-dimensional image of a thin blood vessel is obtained. Can be created accurately. In particular,
Estimate the arrival status of the contrast agent based on information such as the distance from the coronary artery and descending aorta from the sinus of Valsalva, circulation time, heart rate, etc., and correct the maximum CT value to create more accurate images of thin blood vessels It is possible to do.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a medical image creating apparatus according to the present invention.

FIG. 2 is a flowchart illustrating an operation of the medical image creating apparatus of FIG. 1;

FIG. 3 is a diagram illustrating a state of a heart and blood vessels near the heart.

FIG. 4 is a sectional view taken along line AA ′ of FIG. 3;

5 is a long-axis image of a blood vessel 205A at a point C in FIG.

6 is a short-axis image of a blood vessel perpendicular to the long axis of the blood vessel 205A at a point C in FIG. 5;

7 is a profile curve of a blood vessel at a point C in FIG. 5 which is perpendicular to the long axis of the blood vessel 205A.

FIG. 8 is a flowchart illustrating a blood vessel extraction process in FIG. 2;

FIG. 9 is a diagram illustrating an example of a threshold value of a coronary artery set in the transverse images of slice numbers 1 to 40;

FIG. 10 is a diagram showing a correspondence relationship between a CT value and a transparency.

[Explanation of symbols]

101 imaging device, 102 storage device, 103 storage medium, 104 information processing device, 105 input device, 1
06 display device, 201 heart, 202 ascending aorta,
203 sinus Valsalva, 204 right coronary artery, 205 left coronary artery, 205A left anterior descending branch, 205B left circumflex, 2
06 Aortic arch, 207 Descending aorta

Claims (5)

[Claims] 1. A series of cross-sectional images taken by a computer tomography apparatus or the like using a contrast agent to obtain a thin blood vessel lumen
A medical image creating method for creating a three-dimensional image, a measuring step of measuring the maximum pixel value of the aortic lumen located in each cross-section, and a reference value as a parameter to be a reference of extraction of the vessel lumen An extraction step of setting a predetermined ratio of a difference from the maximum pixel value and extracting the blood vessel lumen. 2. An estimation step of estimating an arrival state of a contrast agent based on information on a distance between a thin target blood vessel and the aorta, a distance between the two, a specific site, a circulation time and a heart rate,
2. The medical image creating method according to claim 1, further comprising: a correcting step of correcting the maximum pixel value measured by the measuring means in the arrival situation obtained in the estimating step. 3. A series of cross-sectional images taken by a computer tomographic imaging apparatus or the like using a contrast agent to obtain a thin blood vessel lumen 3
A medical image creating method for creating a three-dimensional image, comprising: a measuring unit that measures a maximum pixel value of an aortic lumen located in each cross section; and a reference value and the maximum pixel as parameters serving as a blood vessel extraction reference. A medical image measuring apparatus, comprising: a predetermined means for setting a predetermined ratio of the difference between the value and the value, and extracting the blood vessel lumen. 4. Estimating means for estimating a state of arrival of a contrast agent based on information on a distance between a thin target blood vessel and the aorta, a distance between the two, a specific site, a circulation time, and a heart rate. A medical image measuring apparatus, comprising: a correcting unit that corrects the maximum pixel value measured by the measuring unit in the estimated arrival situation. 5. A recording medium on which a program for causing a computer to execute the medical image creating method according to claim 1 or 2 is recorded.