JP2007143880A - Determination method and arithmetic unit for contrast medium injection protocol - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a proper injection protocol for monophasic injection without depending on empirical fitting processing. <P>SOLUTION: A contrast medium is test-injected into a subject according to a test injection protocol, and the subject is prescanned (a test: S1-S3). A test time concentration curve is created on the basis of the test (S4 and S5). An injection rate is set at a fixed value equivalent to a test injection rate, according to the test time concentration curve, and a plurality of predicted time concentration curves when the injection rate is determined as various values are created (S7). In each of the predicted time concentration curves, a maximum concentration value capable of securing a required concentration maintenance time is determined (S8). The injection rate is determined on the basis of the ratio of the maximum concentration value to the required concentration value (S9). Additionally, the injection amount of the contrast medium is determined by making the injection rate multiplied by an injection time, and the plurality of injection protocols, which represent the monophasic injection specified by the injection amount and injection rate of the contrast medium, are created (S10). <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a contrast medium injection protocol determination method and a contrast medium injection protocol calculation applicable to an imaging method in which a contrast medium is injected into a subject (typically a human body) to increase the concentration of a region of interest of the subject. Relates to the device. Furthermore, the present invention relates to a contrast medium injection device and an imaging system provided with such a contrast medium injection protocol computing device.

  There is CTA (Computed Tomography Angiography) as means for finding lesions of blood vessels and surrounding organs such as aneurysms and blood vessel occlusions. A CT image obtained by normal CT (Computed Tomography) has a small difference in X-ray absorption values in soft tissues such as organs, blood vessels, and muscles, resulting in a low-contrast image. Therefore, it is difficult to make a satisfactory diagnosis regarding these parts. In contrast, CTA is a technique in which a contrast agent having a high X-ray absorption value is administered to a vein, and CT is continuously performed in a state where the X-ray absorption value of the blood vessel site of interest is increasing. The contrast can be increased and a clear organ image can be obtained.

In general, injection of a contrast agent in CTA is performed by using an iodinated contrast material and determining the total injection amount and injection rate. The enhancement of the image density by the contrast agent is adjusted by changing the delay time from the start of injection to the start of scanning for imaging (Non-patent Document 1).
However, if the injection parameters set based on the injection speed and the injection amount of the contrast agent are not appropriate, the difference in the density value of the image is small, which is not useful for diagnosis, or conversely, the contrast value is too high because the density value is too high. There is a possibility that the image will not be connected. However, the time density change (image density time change) due to contrast medium injection varies greatly depending on the weight of the patient, the disease that the patient has, and the like. Therefore, the dispersion of the concentration value and the delay time from the start of the injection of the contrast agent to the appearance of the contrast agent in the blood vessel in the region of interest is large. Although optimization is a very important matter in obtaining an appropriate CT angiogram, it is not always easy.

Furthermore, in the current medical field, although a contrast medium is a burden on the patient, a slightly larger amount of contrast medium is administered to the patient in order to perform stable contrast. There's a problem.
If the time density curve can be estimated with the aid of a computer for such a problem, it can be expected that the amount of contrast medium to be used can be set appropriately and that it will not be excessively administered. This is thought to reduce the physical and economic burden on the patient.

  As a conventional study for analyzing the time concentration curve, there is a method (Non-patent Document 1) of Freischmann et al. This technique can be applied when a time concentration curve (time change of CT value) when a contrast medium is injected with a test injection parameter for a certain patient is known. That is, in this method, first, Fourier transformation is applied to the test injection parameter and the time concentration curve resulting therefrom, and the relationship between the time concentration change and the patient injection is derived as a transfer function. By using the transfer function, it is possible to estimate the time concentration curve for any injection parameter, and conversely, by setting the time concentration curve required in CT, the optimal injection parameter to obtain it is estimated. can do.

  However, in this method, fitting processing for obtaining the optimal injection parameter is empirically performed. That is, the optimal injection parameter is expressed by a function of a complicated shape in which the injection rate changes nonlinearly with time, but in an actual injection performed using an automatic injector, the injection rate is changed with time. It is unrealistic to make complex changes. Therefore, after performing the so-called two-phase injection, in which the injection speed is switched to two stages, and determining the optimal injection parameters, the fitting process for determining the parameters of the two-phase injection that matches the optimal injection parameters according to the operator's experience. Is done.

In the above method, the two-phase injection is performed because the optimal injection parameter is easy to fit to the two-phase injection, and the fitting is performed for the one-phase injection that keeps the injection speed constant from start to finish. It is because it is not possible. However, there is an aspect that clinical injection is more realistic.
Dominik Freischmann and Karl Hittmair: "Mathematical Analysis of Arterial Enhancement and Optimization of Bolus Geometry for CT Angiography Using the Discrete Fourier Transform", Journal of Computer Assisted Tomography 23 (3): 474-484, 1999 Supervised by Hiroshi Tatsuiri and Kiyoya Inagi, Kazuya Yamashita and Akinori Hayami: “Radiotherapy Technology Volume 10th revised edition”, Nanedo, pp. 83, 86, 175-177, 339, 2001 Supervised by the Japan Society for Medical Image Engineering, edited by the Medical Image Engineering Handbook Editorial Committee, “Medical Image Engineering Handbook”, Shinohara Publishing, pp.220-221, pp.364-365,1994

Accordingly, an object of the present invention is to provide a contrast agent injection protocol determination method and a contrast agent injection protocol calculation device capable of obtaining an appropriate injection protocol for one-phase injection without depending on empirical fitting processing. is there.
Another object of the present invention is to provide a contrast medium injection device using such a contrast medium injection protocol computing device and an imaging apparatus using the same.

  In order to achieve the above object, according to the first aspect of the present invention, there is provided a step (S2) of performing a test injection of a contrast medium over a predetermined test injection time at a constant test injection speed with respect to an imaging target; Steps (S3 to S5) of imaging a region of interest of the target and creating a test time concentration curve representing a temporal change in image density accompanying the test injection of a contrast agent, and the injection speed based on the test time concentration curve Estimating a time concentration curve when a predetermined injection time different from the test injection time is set to a constant value equal to the test injection speed, and creating an estimated time concentration curve (S7); and A step (S8) of obtaining a maximum density value capable of securing a required density maintenance time in the density curve, and based on the maximum density value and the required density value (density enhancement value), A step (S9) of determining an injection rate corresponding to the predetermined injection time so that the required concentration value can be maintained over a predetermined maintenance time, and contrast agent injection based on the predetermined injection time and the injection rate corresponding thereto A method for determining a contrast medium injection protocol, including the step of determining an amount and creating an injection protocol including a pair of the contrast medium injection amount and the corresponding injection speed (S10). The alphanumeric characters in parentheses indicate corresponding components in the embodiments described later. The same applies hereinafter.

  In this method, test injection is performed over a predetermined test injection time at a constant test injection rate. That is, this test injection is a so-called one-phase injection. A test time concentration curve corresponding to the test injection is created. Based on the test time concentration curve, the injection rate is set to a constant value equal to the test injection rate, and the injection time is different from the test injection time. An estimated time concentration curve is created. Further, in the estimated time density curve, a maximum density value that can ensure a required density maintenance time is obtained. Then, based on the maximum density value and the required density value (the density value required at the time of imaging for the actual inspection), an injection speed corresponding to the predetermined injection time is obtained. This injection rate is an injection rate at which a required concentration value can be maintained over the concentration maintenance time. Specifically, it can be obtained by multiplying the test injection rate by the ratio of the required concentration value to the maximum concentration value. Furthermore, a contrast medium injection amount (product of injection time and injection speed) is obtained based on the predetermined injection time and the injection speed corresponding thereto. Thus, a one-phase injection protocol defined by the contrast agent injection amount and the corresponding injection rate can be obtained.

This method does not require empirical fitting in the course of obtaining the injection protocol. Therefore, an appropriate injection protocol that can maintain the required concentration value over the required concentration maintenance time can be reliably created. Moreover, it is possible to create a one-phase injection protocol that can be practically applied in the medical field.
According to a second aspect of the present invention, the predetermined injection time is a natural number multiple of the test injection time, and the step of creating the estimated time concentration curve includes the test time concentration curve and the test time concentration curve as a time. Including a step (S7) of creating an estimated time concentration curve corresponding to the predetermined injection time by superimposing one or more shift time concentration curves shifted on the axis by a natural number multiple of the test injection time. The method for determining a contrast medium injection protocol according to claim 1.

In this method, by limiting the predetermined injection time to a natural number multiple of the test injection time, the estimated time concentration curve can be created by a simple process in which the test time concentration curve is shifted and superimposed on the time axis. .
According to a third aspect of the present invention, the step of creating the estimated time concentration curve is based on the test time concentration curve, and is an impulse response curve (function) that is a time concentration curve corresponding to a predetermined minute unit time contrast agent injection. ), And a step (S26) of obtaining the estimated time concentration curve by shifting a plurality of the impulse response curves on the time axis and superimposing them in accordance with the predetermined injection time. The method for determining a contrast medium injection protocol according to claim 1.

In this method, an impulse response curve representing a response of an imaging target to a contrast medium injection for a predetermined minute unit time is obtained. By multiplying a plurality of impulse response curves shifted on the time axis, an estimated time concentration curve corresponding to an arbitrary injection time can be created.
According to a fourth aspect of the present invention, in the step of obtaining the impulse response curve, a step (S21) of performing a Fourier transform on a test injection function representing a time change of an injection speed at the time of test injection, and a Fourier transform of the test time concentration curve. Step (S22) and a step of obtaining a transfer function representing the relationship between the injection function representing the time change of the injection rate and the time concentration curve by taking the ratio of each Fourier transform of the test injection function and the test time concentration curve ( 4. The contrast medium injection protocol determination method according to claim 3, comprising: S24) and a step of obtaining the impulse response curve by performing inverse Fourier transform on the transfer function (S25).

In this method, a transfer function representing the relationship between the injection function and the time concentration curve is obtained by taking the ratio of each Fourier transform of the test injection function and the test time concentration curve. The impulse response curve can be obtained by inverse Fourier transforming this transfer function.
According to a fifth aspect of the present invention, in the step of creating the estimated time concentration curve, a step (S21) of Fourier transforming a test injection function representing a time change of an injection speed at the time of test injection, and a step of converting the test time concentration curve to Fourier The transfer function (S24) representing the relationship between the injection function representing the time change of the injection speed and the time concentration curve by taking the ratio of each Fourier transform of the step (S22) for conversion and the test injection function and the test time concentration curve. And a step of Fourier transforming a set injection function which is an injection function when the injection rate is set to a constant value equal to the test injection rate and the injection time is set to the predetermined injection time (S31, S32). And an estimated time concentration curve corresponding to the set injection function based on the Fourier transform of the set injection function and the transfer function. To create a step (S33, S34) and a is a contrast injection protocol determination process of claim 1.

  In this method, the Fourier transform of the test injection function and the test time concentration curve is obtained, and the Fourier transform of the set injection function, which is an injection function when the injection time is set to a predetermined injection time different from the test injection time, is obtained. . Then, the transfer function is obtained by taking the ratio of each Fourier transform of the test injection function and the test time concentration curve. Furthermore, an estimated time concentration curve corresponding to the set injection function is created based on the transfer function and the Fourier transform of the set injection function. Since the transfer function represents the relationship of the time concentration curve to an arbitrary injection function, the time concentration curve can be estimated by using this transfer function. More specifically, an estimated time concentration curve in the frequency domain is created based on the Fourier transform of the transfer function and the set injection function (S33), and further inverse Fourier transformed to generate an estimated time concentration curve in the time domain. Can be obtained (S34).

The invention according to claim 6 further includes a step (S23) of performing a filtering process for removing a high frequency component on a Fourier transform of the test time density curve, wherein the step of obtaining the transfer function includes the filtered test. 6. The method for determining a contrast medium injection protocol according to claim 4 or 5, which is performed using Fourier transform of a time density curve.
In this method, filter processing for removing high-frequency components is performed on the Fourier transform of the test time density curve. Thus, when combined with the invention according to claim 4, it is possible to create a good impulse response curve that does not include a high-frequency component. Further, when combined with the invention described in claim 5, a transfer function that does not include a high-frequency component can be obtained, so that a good estimated time concentration curve can be obtained.

In the invention of claim 7, the predetermined injection time includes a plurality of different injection times,
The step of creating the estimated time concentration curve is a step of creating a plurality of estimated time concentration curves respectively corresponding to the plurality of injection times, and the step of obtaining a maximum concentration value capable of securing the concentration maintenance time is the plurality of steps. Each of the estimated time concentration curves is a step of obtaining a maximum concentration value capable of securing the concentration maintenance time, and the step of obtaining the injection rate includes a plurality of maximum concentrations obtained for each of the plurality of estimated time concentration curves. Obtaining a plurality of injection speeds corresponding to the plurality of injection times based on the value and the required concentration value, wherein the step of creating the injection protocol corresponds to the plurality of injection speeds and to each of them. Creating a plurality of injection protocols each including a plurality of pairs of a plurality of contrast agent injection amounts There is a contrast injection protocol determination method according to any one of claims 1 to 6.

By this method, injection speeds corresponding to a plurality of injection times are obtained, and contrast agent injection amounts corresponding to these are obtained. As a result, a plurality of injection protocols each including a set of contrast agent injection amount and injection speed can be obtained. An appropriate injection protocol may be selected from the plurality of injection protocols in consideration of the injection speed and the contrast agent injection amount.
The invention according to claim 8 is the method for determining a contrast medium injection protocol according to claim 7, further comprising a step (S12) of extracting an injection protocol whose injection rate is a value within a predetermined range from the plurality of injection protocols. is there.

In this method, since an injection protocol having an injection rate within a predetermined range can be extracted, the plurality of injection protocols can be further narrowed down to actually applicable injection protocols.
The invention according to claim 9 is the method for determining a contrast medium injection protocol according to claim 7 or 8, further comprising a step (S12) of extracting an injection protocol having the lowest injection rate from the plurality of injection protocols.

According to this method, since an injection protocol having the lowest injection speed is extracted, it is possible to select an injection protocol with a small burden on the imaging target (patient).
The invention according to claim 10 further includes a step (S12) of extracting an injection protocol having the lowest contrast agent injection amount from the plurality of injection protocols (S12). Protocol determination method.

According to this method, since the injection protocol with the lowest contrast agent injection amount is extracted, it is possible to select an injection protocol with less burden on the imaging target (patient).
The invention according to claim 11 is based on a test time density curve obtained by imaging a region of interest when a contrast medium is test-injected over a predetermined test injection time at a constant test injection speed with respect to an imaging target. An apparatus (2) for calculating a contrast medium injection protocol for an actual examination, wherein an injection rate is set to a constant value equal to the test injection rate based on the test time concentration curve, and is different from the test injection time. Estimated time concentration curve creating means (37) for estimating the time concentration curve when the injection time is set and creating the estimated time concentration curve, and the estimated time concentration created by the estimated time concentration curve creating means Maximum density value calculation means (38) for obtaining a maximum density value capable of securing a required density maintenance time in the curve, and before calculation by this maximum density value calculation means An injection rate calculating means (39) for determining an injection rate corresponding to the predetermined injection time so that the required concentration value can be maintained over the concentration maintenance time based on the maximum concentration value and the required concentration value; A contrast medium injection amount is obtained based on the injection time and the injection speed calculated by the injection speed calculation means correspondingly, and an injection protocol including a pair of the contrast medium injection amount and the corresponding injection speed is created. A contrast medium injection protocol computing device including an injection protocol creating means (40).

According to this configuration, it is possible to automatically obtain an appropriate one-phase injection protocol that can ensure a required concentration value over a predetermined concentration maintaining time based on the result of the test injection.
Of course, the invention of the contrast medium injection protocol computing device can be modified in the same manner as described in connection with the above-described invention of the contrast medium injection protocol determination method.
The above-described contrast medium injection protocol computing device can be realized by causing a computer to execute a predetermined computer program. In this case, the computer program is executed on the computer to cause the computer to function as each of the function realizing means. Such a computer program may be provided in the form of a recording medium readable by a computer, or may be provided in the form of a signal propagating through a communication line.

  According to the invention of claim 12, the contrast medium is injected into the imaging object according to the contrast medium injection means (3) for injecting the contrast medium into the imaging object and the test injection protocol corresponding to the test injection speed and the test injection time. In accordance with the test injection control means (32, S1, S2) for controlling the contrast agent injection means, the contrast agent injection protocol calculation device according to claim 11, and the injection protocol calculated by the contrast agent injection protocol calculation device A contrast medium injecting apparatus including actual injection control means (32, S13) for controlling the contrast medium injecting means so that the contrast medium is injected into the imaging object.

With this configuration, an appropriate one-phase injection protocol is obtained based on the result of the test injection, and a contrast agent can be injected into the imaging target according to the obtained injection protocol.
A thirteenth aspect of the present invention is the contrast agent injecting device according to the twelfth aspect of the present invention, the photographing means (1) for photographing the photographing target and generating a signal corresponding to the image density, and the region of interest of the photographing target during the test injection When the contrast medium injecting means is controlled by the test imaging control means (31, S3) for controlling the imaging means so as to take an image, and the actual injection control means, the interest of the imaging object is synchronized with this. And an actual photographing control means (31, S13) for controlling the photographing means so as to photograph a region.

  With this configuration, the test inspection and the actual inspection can be automatically performed by controlling the contrast medium injection device and the imaging device and applying an appropriate one-phase injection protocol.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
1. Configuration of Shooting System FIG. 1 is a block diagram showing a configuration example of a shooting system according to an embodiment of the present invention. This imaging system includes a CT scanner 1, an operation console 2, and an automatic injector 3. The CT scanner 1 is an apparatus for taking a cross-sectional image of a subject (a patient in a medical field). The automatic injector 3 is a device for automatically injecting an angiographic contrast agent into a subject. The operation console 2 exchanges data between the CT scanner 1 and the automatic injector 3 and controls them. More specifically, the operation console 2 gives the CT scanner 1 control data such as a scan range, a scan start time, and the like, and receives data representing the imaging result from the CT scanner 1. Further, the operation console 2 sets an injection protocol (including a contrast medium injection amount and an injection speed in this embodiment) for the automatic injector 3, and operation information from the automatic injector 3 (injection start, End of injection).

  The operation console 2 has a basic configuration as a computer. More specifically, a display device 5 for displaying a two-dimensional image, a keyboard, a mouse and other operation input unit 6, a data processing unit 7 including a CPU, a storage unit 8 for recording various data and programs, An interface unit 9 for exchanging data with respect to the CT scanner 1 and the automatic injector 3 is provided. The storage unit 8 is composed of an appropriate storage medium such as a RAM, a ROM, a hard disk, and the like, and at least a part of the storage area can be written with data.

The data processing unit 7 operates according to a program stored in the storage unit 8, whereby the data processing unit 7, the storage unit 8, and the like constitute the contrast medium injection protocol arithmetic device according to this embodiment. Further, the contrast medium injection protocol computing device and the automatic injector 3 constitute the contrast medium injection apparatus according to this embodiment.
The automatic injector 3 exchanges data between the operation console 2 and an injection head 10 for injecting a contrast medium into a blood vessel of a subject, a control unit 11 for controlling the operation of the injection head 10, and the operation console 2. And an interface unit 12 for this purpose. The control unit 11 controls the injection head 10 according to the injection protocol delivered from the operation console 2. In addition, the control unit 11 monitors the operating state of the injection head 10 and notifies the operation console 2 of the start and completion of contrast agent injection.
2. CT (Computed Tomography)
A CT image is a cross-sectional image of a subject and can be obtained by performing image reconstruction processing on scan data obtained using X-rays.
2-1. FIG. 2 shows an example of a scanning method by the CT scanner 1. FIG. 2 shows a scanning method based on the third generation rotate / rotate method, in which fan-shaped X-rays 16 (fan beam method) having a spread angle of 30 to 40 ° including the subject 15 are generated from the X-ray tube 17. A Xe ionization chamber detector having 500 to 900 detection channels on an arc is arranged as a detector 18 at a position facing the X-ray tube 17. The X-ray tube 17 and the detector 18 rotate together to scan the subject 15 (see Non-Patent Document 2).

  The X-ray 16 emitted from the X-ray tube 17 passes through the subject 15, is detected by the detector 18, and its intensity is measured as an electric signal. The X-ray 16 is absorbed while passing through the subject 15, and its intensity is weakened. The degree of X-ray 16 absorption (X-ray attenuation coefficient) varies depending on the substance. Therefore, in CT, the subject 15 is represented by the spatial distribution f (x, y, z) of the X-ray attenuation coefficient.

The X-ray tube 17 and the detector 18 rotate on z = constant plane (a plane orthogonal to the body axis of the subject 15), and measure the intensity of transmitted X-rays along all straight lines included in this plane. To do. Thereby, data related to a two-dimensional distribution of f (x, y, z) {z = constant} is collected. This is called scanning, and the obtained data is called scan data (transmission X-ray data).
The scan data is a collection of one-dimensional projection distributions of the subject 15. This data is transferred from the CT scanner 1 to the operation console 2, and image reconstruction is performed in the operation console 2. As a result, a two-dimensional image representing the distribution of the X-ray attenuation coefficient is obtained. If the same thing is repeated by changing the slice plane z little by little, the three-dimensional distribution f (x, y, z) is finally expressed as a set of many two-dimensional images.

As a method for obtaining three-dimensional data, helical CT is exemplified. In this method, the X-ray tube 17 and the detector 18 are continuously rotated while irradiating the X-ray 16, and at the same time, a bed (not shown) on which the subject 15 is placed is moved in the body axis direction, and the two motions are combined. Projection data obtained by spirally scanning the subject 15 is obtained.
Since many human organs have an X-ray attenuation coefficient close to that of water, it is convenient to use a scale based on water. Therefore, the value of the X-ray attenuation coefficient is converted so that water becomes 0 and air becomes −1000. This scale is referred to as a CT value (unit: Hounsfield Unit = HU) (Non-patent Document 3).
2-2. Image reconstruction in CT Image reconstruction in CT is to reproduce the distribution of physical quantities related to the radiation source as a two-dimensional image from the one-dimensional projection distribution of the subject obtained from multiple directions. There are several mathematical methods for reconstructing an image from projection data according to Radon's principle, which are roughly divided into a back projection method, a successive approximation method, and an analytical method.

  As shown in FIG. 3, an intuitive method for estimating the distribution of attenuation coefficients from transmission X-ray data (projection data) is the back projection method. When the transmission X-ray data values obtained by transmitting through the object are projected in reverse and the back projections in all directions are superimposed, the attenuation coefficient of the original position can be estimated. Actually, the transmission X-ray data contains “blur” and cannot be used as it is. However, basically, the correction of the transmission X-ray data and the process of backprojecting it are performed mathematically. ing.

In X-ray CT, a filter-corrected back projection method or a convolution back projection method is generally used. Specific procedures of the filtered back projection method and the convolution back projection method are shown below.
As shown in FIG. 4, an object is placed on a stationary orthogonal coordinate system (x, y), and the distribution of the X-ray attenuation coefficient of this object on the (x, y) plane is represented by f (x, y). Shall be represented. A new coordinate system (s, t) inclined by an angle θ with respect to the (x, y) coordinate system is defined, and X-rays are irradiated from the outside of the object along the t-axis to be incident on the object. To do. At this time, the projection data p (s, θ) (transmission X-ray data) is obtained by performing line integration along a straight line L _s, θ (a straight line parallel to the t-axis and displaced by s from the coordinate system origin in the s-axis direction). And is represented by the following formula (1).

  Reconstructing the distribution f (x, y) from projection data p (s, θ) in all directions obtained by gradually changing θ is image reconstruction. Here, the two-dimensional Fourier transform of the image f (x, y) is as in the following equation (2), and the one-dimensional Fourier transform for s of the projection data p (s, θ) is as in the following equation (3). is there.

  Here, since the coordinate obtained by rotating (x, y) by the angle θ is (s, t), the equation (1) can be rewritten as the following equation (4).

  Substituting equation (4) into equation (3) and variable-converting (s, t) into (x, y), the following equation (5) is obtained.

  When Expression (2) and Expression (5) are compared, the relationship of the following Expression (6) is obtained.

This equation expresses a two-dimensional Fourier transform F (u, u) where the one-dimensional Fourier transform P (w, θ) of the projection data p (s, θ) on the s-axis of f (x, y) is f (x, y). It is shown that it matches the section F (wcosθ, wsinθ) on the w-axis of v). This is called the central section theorem, and image reconstruction is performed using this property.
The inverse Fourier transform for obtaining f (x, y) from F (u, v) is as shown in the following equation (7).

  In this equation (7), when variable transformation u = wcos θ and v = wsin θ are performed, the following equation (8) is obtained.

  In this equation (8), the integral with respect to w represents the inverse Fourier transform of the product of P (w, θ) and | w |, which is the inverse Fourier transform p (s, θ) and | w | can be calculated as convolution of inverse Fourier transform. Accordingly, when the inverse Fourier transform of | w | is q (s), the following equation (9) is obtained. q (s) is called a correction function and is represented by the following equation (10).

In equation (9), if the correction function q (s) is convolved with the projection data p (s, θ), and the convolution value for s satisfying s = x cos θ + y sin θ is added in the angular direction, f (x, y) is obtained. Indicates that reconfiguration is possible. Expression (8) in which the filter correction process is performed by frequency domain multiplication is called a filter-corrected back projection method, and Expression (9) in which spatial domain convolution is performed is called a convolution back projection method.
2-3. CTAgiography (CTA; computed tomography angiography)
CTA is CT performed by administering a contrast medium to a blood vessel, and is an imaging method performed for the purpose of examining hemodynamics or diagnosing vascular lesions.

  In normal X-ray photography, soft tissues such as organs, blood vessels, and muscles have small differences in X-ray absorption values, making it difficult to perform satisfactory diagnosis. Therefore, by directly injecting the X-ray contrast medium into the blood vessel, a clear image of the organ can be obtained due to the difference between the X-ray absorption values of the contrast medium and the human tissue. Therefore, the contrast of organs and blood vessels is more emphasized in the CTA image than in a normal CT image. In abdominal angiography (for example, angiography of an abdominal aorta), a water-soluble iodine contrast agent having an X-ray absorption value higher than that of soft tissue is used as a contrast agent (Non-patent Document 2).

On the other hand, since the contrast agent is carried in the blood vessel with the passage of time, the density value of each part changes with the passage of time in the grayscale image obtained by injecting the contrast agent. A curve representing a temporal change in density value in a region of interest (region of interest) in a CT contrast image is called a “time density curve”.
3. 3. Analysis of time concentration curve using Fourier transform 3-1. Target data When performing CT imaging of the abdominal aorta in clinical practice, first, prior to the actual examination, a test injection is performed using 10 to 20 ml of contrast medium as a test examination, and a test time concentration curve (CT value of the abdominal aorta) A time concentration curve representing a change with time) is obtained as data. Based on the data, the contrast agent arrival time to the region of interest is examined, and an appropriate CT scan imaging start time is determined, and then imaging for the actual examination is performed.

Conventionally, the test time density curve is measured only for the purpose of determining the start time of the CT scan of the actual inspection. On the other hand, in this embodiment, by using two data of the test time concentration curve and the test injection parameter, the characteristics of the time concentration curve for each patient (subject) are analyzed, and the time in the actual examination is analyzed. Estimate concentration curve data.
Therefore, in this embodiment, an object to be analyzed and estimated is obtained by performing time series data (injection parameters) of contrast medium injection setting in CTA and injection thereof as shown in FIGS. 5 (a) and 5 (b). It is the time-series data (time density curve) of the obtained density change.

FIG. 5 (a) shows time series data of contrast medium injection speed set in the automatic injector 3 (see FIG. 1). This is hereinafter referred to as “contrast medium injection function” or simply “injection function” (contrast medium injection). It is called a function that represents a change in speed over time. When distinguishing the contrast medium injection function at the time of the test examination, it is called a “test injection function”.
FIG. 5B is an example of time-series data of density values of the abdominal aorta portion of the contrast image acquired by the CTA described in Section 2-3. Such time-series data represents a “time density curve”. Become. During the test examination, CTA is continuously performed for a specific time on a specific cross section of the abdomen, and a contrast image is obtained at regular intervals within that time. When distinguishing the time density curve obtained by the test inspection, it will be called a “test time density curve”.
3-2. Basic Theory for Analysis In the conventional method (Non-patent Document 1), as shown in FIG. 6, the complex system of the human body is regarded as a black box, and only the input and output are analyzed, that is, the contrast medium is injected into the device. The analysis is performed using only two data of the function inj and the time concentration curve enh obtained thereby.

The contrast agent injection function inj is time series data obtained from two values of “injection speed” and “total injection amount” which are set values of the automatic injector 3 (contrast agent injection device). Think of it as an input. The intermediate process from input to output is represented by one transfer function, which is hereinafter referred to as “transfer function G”.
The transfer function G is derived using a contrast agent injection function (test injection function) of a subject obtained by performing a test in advance and a value of a time density curve (test time density curve) obtained thereby. By using this transfer function G, a time density curve corresponding to an arbitrary contrast agent injection function can be estimated for the subject (patient). Conversely, it is possible to estimate a contrast agent injection function for obtaining an ideal time density curve.

  Assuming that the black box is a linear time-invariant system, the arterial time-concentration curve enh is expressed as follows: A system function g representing the system can be obtained by convolution and integration, and can be analyzed using Fourier transform.

3-3. Analysis and estimation using Fourier transform Equation (11) expresses a mathematical relationship for analysis in the time domain. Here, the arterial time density curve enh is a convolution of two functions of the contrast agent injection function inj and the system function g, and therefore, when the Fourier transform is applied to the equation (11), it is expressed as the following equation (12). be able to.

This is a calculation in the frequency domain, and Enh, Inj, and G are functions in the frequency domain obtained by Fourier transforming each function of Equation (11).
The derivation process of the transfer function G including the time domain calculation is shown in FIG.
The transfer function G is obtained using the following equation (13) by measuring in advance the test injection function Inj _test of the contrast agent and the test time concentration curve Enh _test obtained thereby.

Once the transfer function G is obtained, it is easy to estimate the time density curve Enh from the contrast agent injection function Inj.
The time density curve estimation process is shown in FIG.
A production time concentration curve Enh _std obtained when a certain production injection function Inj _std is given as an input can be estimated by the following equation (14).

If inverse Fourier transform is applied to Enh _std obtained by Expression (14), a time concentration curve enh _{std in the} time domain is obtained.
Here, from the characteristics of the Fourier transform, an ideal injection function Inj _ideal for obtaining an _ideal time concentration curve Enh _ideal can be calculated by the following equation (15). FIG. 9 shows an estimation process of the _ideal injection function Inj _ideal .

In the frequency spectrum of the transfer function G, Equation (15) is effective for non-zero frequency components. However, in general, since the high-frequency component of the transfer function G includes a value of 0, it should be noted that the corresponding Enh _ideal high-frequency component should not be calculated.
The theoretical value inj _ideal of the _ideal contrast agent injection function for obtaining the ideal time density curve of the abdominal aorta is calculated by inverse Fourier transforming the injection function Inj _ideal obtained by equation (15) into the time domain. be able to.

For clinical use, it is necessary to perform fitting to the ideal value of an ideal contrast agent injection function and determine injection parameters that do not cause any problems even if injection is performed on the human body.
As shown in Non-Patent Document 1, in order to fit a realistic contrast agent injection function with an ideal contrast agent injection function, two-phase injection is required. Often, biphasic injection is not practical. Therefore, actually, even if an ideal contrast agent injection function is obtained by Equation (15), there is no practical means for realizing this.
4). Estimation of Contrast Agent Injection Parameter FIG. 10 is a block diagram for explaining a functional configuration of the data processing unit 7 (see FIG. 1) provided in the operation console 2. The data processing unit 7 implements functions as a plurality of function processing units shown in FIG. 10 by executing a predetermined program. These function processing units include a scan control unit 31 for controlling the CT scanner 1, an injection control unit 32 for controlling the automatic injector 3, and a subject based on transmission X-ray data collected by the CT scanner 1. An image reconstructing unit 33 for reconstructing the cross-sectional image, and an injection protocol calculating unit 34 for calculating a contrast agent injection protocol at the time of the actual injection based on data obtained by the test injection.

  The injection control unit 32 reads the test injection protocol from the storage unit 8 and transfers it to the control unit 11 of the automatic injector 3 via the interface unit 9, and the operation information (contrast image) of the automatic injector 3 from the control unit 11. Information on the start and end of injection of the agent is received via the interface unit 9. The test injection protocol may be configured so that the operator inputs in advance by operating the operation input unit 6, and the input test injection protocol is stored in the storage unit 8. This test injection protocol, in this embodiment, is an injection protocol corresponding to a single phase injection, and is represented by a set of test contrast agent injection volume and test injection rate. In this case, the test injection function is a rectangular function in which a constant test injection speed continues for a test injection time (= test contrast medium injection amount / test injection speed).

  The scan control unit 31 controls the CT scanner 1 in cooperation with the injection control unit 32. That is, information on the start and end of contrast agent injection is obtained from the injection controller 32, and the start and end of the imaging operation by the CT scanner 1 are controlled. During this period, the CT scanner 1 continues to capture a cross section including the region of interest of the subject and outputs transmitted X-ray data. The transmitted X-ray data is received via the interface unit 9 and accumulated in the storage unit 8.

The image reconstruction unit 33 reconstructs a two-dimensional image based on the transmission X-ray data accumulated in the storage unit 8. Thereby, a plurality of cross-sectional images according to time series are reconstructed, and image data representing the plurality of cross-sectional images is written in the storage unit 8.
The injection protocol calculator 34 includes a test time concentration curve generator 36, an estimated time concentration curve generator 37, a maximum concentration value calculator 38, an injection rate calculator 39, an injection protocol generator 40, and an injection protocol selection. The injection protocol used in the actual inspection is calculated.

The test time density curve creation unit 36 pays attention to a region of interest in a plurality of cross-sectional images created by the image reconstruction unit 33 at the time of test injection, and represents a test time density curve ( Image density time series data). This test time density curve is stored in the storage unit 8.
The estimated time concentration curve creation unit 37 reads the test time concentration curve from the storage unit 8 and corresponds to a plurality of contrast agent injection times (injection time candidates) assumed in the actual examination based on the test time concentration curve. Create an estimated time concentration curve. This estimated time concentration curve is a time concentration curve that is expected to be observed in the region of interest when the contrast agent injection rate is set equal to that at the time of test injection and the contrast agent is injected into the subject at various injection times. It is. The plurality of estimated time concentration curves created in this way are stored in the storage unit 8.

  The maximum density value calculation unit 38 reads the plurality of estimated time density curves from the storage unit 8 and obtains the maximum density value h that can satisfy the predetermined density maintenance time T in each estimated time density curve. The density maintenance time T may be set by the operator through a setting operation from the operation input unit 6. This density maintenance time T is a time during which the image density of the region of interest should be maintained at a predetermined required density value H (density enhancement value) or more during CTA imaging. The maximum density value h obtained for each estimated time density curve is stored in the storage unit 8.

  The injection rate calculation unit 39 reads the maximum concentration value h obtained for each estimated time concentration curve from the storage unit 8 and compares each maximum concentration value h with the required concentration value H. Based on the comparison result, the injection rate calculation unit 39 calculates the contrast agent injection rate v that can achieve the required concentration value H for each estimated time concentration curve (that is, for each injection time). To do. The plurality of calculated contrast medium injection speeds v are stored in the storage unit 8 in association with the respective injection times.

The injection protocol creation unit 40 reads the calculated plurality of injection speeds v from the storage unit 8 and multiplies them by the corresponding injection speeds to obtain the contrast agent injection amount (= injection speed × injection time). In this way, a plurality of contrast agent injection protocols including a set of contrast agent injection amount and injection speed v are created and stored in the storage unit 8.
The injection protocol selection unit 41 selects an appropriate contrast agent injection protocol from among the plurality of contrast agent injection protocols stored in the storage unit 8 based on the contrast agent injection conditions set by the input from the operation input unit 6. This is selected and stored in the storage unit 8 as the actual injection protocol (injection protocol to be used for the actual inspection). This actual injection protocol is transferred to the control unit 11 of the automatic injector 3 by the injection controller 32.
4-1. Concentration and generation of concentration curve using Fourier transform Assuming that the complex system of human body is a linear time-invariant system, the injection rate is set to a constant value equal to the test injection rate (v _test ), and the injection time is set to As shown in FIG. 11, the time concentration curve when the test injection time t _{test is} a natural number multiple can be directly calculated in the time domain by superimposing the test time concentration curves. More specifically, a shift time concentration curve 22 in which the test time concentration curve 21 is shifted in the time axis direction by the test injection time t _test is created, and these curves 21 and 22 are overlapped to obtain 2 of the test injection time 2. An estimated time concentration curve 23 corresponding to the double injection time can be created. Similarly, it is possible to create an estimated time concentration curve corresponding to an injection time that is an arbitrary natural number times the test injection time.

FIG. 12 is a block diagram for explaining a configuration example of the estimated time concentration curve creation unit 37. The estimated time concentration curve creation unit 37 creates a shift time concentration curve by shifting the test time concentration curve read from the storage unit 8 in the time axis direction by a natural number multiple of the test injection time t _test , And an overlay processing unit 46. The overlay processing unit 46 performs overlay processing for overlaying the shift time density curve created by the shift processing unit 45 and the test time density curve read from the storage unit 8.

When the shift processing unit 45 creates an estimated time concentration curve corresponding to an injection time twice as long as the test injection time t _test , the shift processing unit 45 shifts the test time concentration curve in the time axis direction by the test injection time t _test. Create a concentration curve. When an estimated time concentration curve corresponding to an injection time three times the test injection time is to be created, a first shift time concentration curve in which the test time concentration curve is shifted in the time axis direction by the test injection time t _test , A second shift time concentration curve is created by shifting the test time concentration curve in the time axis direction by a time twice as long as the test injection time _ttest . When creating an estimated time concentration curve corresponding to an injection time that is a natural number multiple of four times or more of the test injection time, three or more shift time concentration curves that are shifted on the time axis by the test injection time t _test accordingly. Is generated by the shift processing unit 45. One or a plurality of shift time concentration curves thus created by the shift processing unit 45 are superimposed on the test time concentration curve by the superimposition processing unit 46, so that an arbitrary natural number of times of the test injection time _ttest is injected. An estimated time concentration curve corresponding to time can be created.

In the time domain calculation, there is a restriction that the concentration curve that can be generated is limited to an injection time that is a natural number multiple of the test injection time.
Therefore, by using the Fourier transform, an estimated time concentration curve for an arbitrary injection function can be obtained from the relationship of FIG. 7 and FIG. 8, so that the injection time was changed to an arbitrary value under the condition of a constant injection speed. An estimated time concentration curve can be generated.

FIG. 13 is a block diagram for explaining another configuration example of the estimated time concentration curve creating unit 37. The estimated time concentration curve creation unit 37 includes a first Fourier transform unit 51 that performs a Fourier transform on the test injection function inj _test , a second Fourier transform unit 52 that performs a Fourier transform on the test time concentration curve enh _test , and a second Fourier transform unit. 52, the filter processing unit 54 that performs a filtering process to remove high frequency components on the frequency domain test time density curve Enh _inj , and the calculation results of the first Fourier transform unit 51 and the filter processing unit 54 ( Based on the frequency domain test injection function Inj _test and the test time concentration curve Enh _test (after filtering)), the transfer function calculation unit 53 that calculates the transfer function G according to the above equation (13), and the transfer function calculation An inverse Fourier transform unit 55 that performs an inverse Fourier transform on the transfer function G obtained by the unit 53 to obtain an impulse response; The shift processing unit 57 that shifts the response on the time axis, and the impulse response shifted by the shift processing unit 57 are superimposed on the impulse response obtained by the inverse Fourier transform unit 55 to thereby estimate the estimated time concentration of an arbitrary injection time. It includes an overlay processing unit 56 for calculating a curve.

  The inverse Fourier transform of the transfer function G becomes a system function g. This system function g corresponds to an impulse response (for example, a minute unit time (for example, 0.25 seconds)) of a black box (see FIG. 6) corresponding to the subject. Specimen response). Therefore, an estimated time concentration curve corresponding to an arbitrary injection time can be obtained by shifting this impulse response and shifting the impulse response on the time axis in accordance with the injection time for which an estimated time concentration curve is desired to be obtained.

For example, the filter processing unit 54 removes high frequency components by removing frequency components less than a predetermined ratio (for example, 1%) with respect to the maximum amplitude in the spectrum of the test time concentration curve. Thereby, it can be avoided that the high-frequency component of the transfer function G includes a zero value.
FIG. 14 (a) shows an impulse response (system function g) when the transfer function G is obtained from the test time concentration curve in the frequency domain without removing the high frequency component, and FIG. 14 (b) shows the above-described high frequency elimination filter. The impulse response (system function g) when the transfer function G is obtained based on the processed test time concentration curve is shown. From these comparisons, it can be seen that high-frequency components of the impulse response can be removed by filtering.

15 (a) to 15 (h) are superimposed on the impulse response of FIG. 14 (b) for various injection times (1 second, 2 seconds, 3 seconds, 4 seconds, 5 seconds, 10 seconds, 15 seconds, 20 seconds). The example which produced | generated the estimation time density | concentration curve of (second) was shown. In this way, it is possible to generate an estimated time concentration curve for an arbitrary injection time, not limited to an injection time which is a natural number times the test injection time.
FIG. 16 shows an example (solid line) of the estimated time concentration curve created by the configuration of FIG. 12 and an example (broken line) of the estimated time concentration curve created by the configuration of FIG. These were generated from the same test time concentration curve.

  From the comparison of the two curves in FIG. 16, even when some high-frequency components are excluded by the filter processing, an estimated time concentration curve that is almost the same as the time concentration curve generated by superposition can be obtained. You can see that it is very small. Therefore, by using the Fourier transform, it is possible to satisfactorily estimate a time concentration curve for an arbitrary injection time while avoiding restrictions on the injection time. On the contrary, when the protocol calculation for the injection time that is a natural number multiple of the test injection time is sufficient, the time concentration curve can be satisfactorily estimated by a simple process including no Fourier transform.

FIG. 17 is a block diagram for explaining still another configuration example of the estimated time concentration curve creating unit 37. In FIG. 17, the same functional parts as those shown in FIG. 13 are given the same reference numerals as in FIG. The estimated time concentration curve creation unit 37 includes a first Fourier transform unit 51 that performs a Fourier transform on the test injection function inj _test , a second Fourier transform unit 52 that performs a Fourier transform on the test time concentration curve enh _test , and a second Fourier transform unit. 52, the filter processing unit 54 that performs a filtering process for removing high frequency components on the frequency domain test time density curve Enh _test , and the calculation results of the first Fourier transform unit 51 and the filter processing unit 54 ( Based on the frequency domain test injection function Inj _test and the test time concentration curve Enh _test (after filtering)), a transfer function calculation unit 53 that calculates the transfer function G according to the equation (13), Is a third Fourier transform that creates a Fourier transform of an injection function (set injection function: time-series data of injection speed) corresponding to a plurality of different injection times. The estimated time concentration curve Enh in the frequency domain is obtained according to the above equation (14) using the calculation result (frequency domain injection function) of the unit 58 and the third Fourier transform unit 58 and the transfer function G after the filtering process. An estimated time concentration curve calculating unit 59 and an inverse Fourier transform unit 60 for obtaining an estimated time concentration curve enh in the time domain by performing inverse Fourier transform on the calculation result of the estimated time concentration curve calculating unit 59 are provided. Yes.

  With this configuration, a time concentration curve (estimated time concentration curve) to be obtained by an injection function corresponding to an arbitrary injection time can be calculated. However, in this embodiment, since the contrast medium is injected by one-phase injection, the injection function given to the third Fourier transform unit 58 has the injection speed equal to the test injection speed, and only the injection time is the test injection. A rectangular function different from time. Therefore, in general, the process in the case of the configuration of FIG. 13 is simpler.

Since the high-frequency component of the test time concentration curve Enh _test is previously removed by the function of the filter processing unit 54, it is possible to avoid including a zero value in the high-frequency component of the transfer function G.
4-2. Estimation of Contrast Agent Injection Parameter FIG. 18 is a flowchart for explaining a processing procedure for obtaining an injection protocol for the actual examination when the estimated time concentration curve creation unit 37 has the configuration of FIG. 12 described above. The operator inputs a test injection protocol by operating the operation input unit 6 in advance (step S1). This test injection protocol includes, for example, a test injection rate v _test , a test injection time t _test and a test injection amount a _test , and a = v _test t (a is an injection amount, v is an injection rate, and t is a time) That is, it is determined that the contrast medium is injected at a constant test injection speed v _test over the test injection time. In this case, the test injection function inj _test is represented by a rectangular function in which a constant test injection speed v _test continues for the test injection time t _test .

  The injection control unit 32 passes the test injection protocol to the control unit 11 of the automatic injector 3. Thereby, the automatic injector 3 performs test injection of the contrast agent to the subject according to the test injection protocol (step S2). Operation information representing the operation of the automatic injector 3 at this time is notified from the control unit 11 to the injection controller 32, and the scan controller 31 controls the CT scanner 1 based on this operation information. In this way, in parallel with the test injection of the contrast agent, the subject is pre-scanned (scan for test inspection) (step S3).

  The transmitted X-ray data collected by the CT scanner 1 is stored in the storage unit 8 of the operation console 2, and based on the transmitted X-ray data, the image reconstruction unit 33 represents a plurality of time-series CTA contrast images. Image data is created (step S4). Further, a test time density curve is created by the test time density curve creating unit 36 based on the plurality of pieces of image data (step S5).

Next, the operator inputs from the operation input unit 6 a concentration enhancement value HU [ΔHU] and a maintenance time T (seconds) desired to be obtained during the actual inspection (step S6).
As described above, the estimated time concentration curve creation unit 37 makes the injection rate constant at a value equal to the test injection rate and sets the injection time to be a natural number multiple of the test injection time by superimposing the test time concentration curves. The estimated value of the time density curve to be obtained (estimated time density curve) is created and stored in the storage unit 8 (step S7). The estimated time concentration curve creation unit 37 performs the same processing for a plurality of different injection times, and creates a plurality of estimated time concentration curves.

Examples of the estimated time concentration curve created are shown in FIGS. 19 (a), 19 (b), and 19 (c). The upper part of each figure shows the injection function inj (function representing the time change of the injection speed v), and the estimated time concentration curve enh is shown at the lower part of each figure.
FIG. 19A shows the test injection function inj ₁ and the test time concentration curve enh ₁ corresponding thereto. In this example, the test injection v _test speed is 3 ml / second and the test injection time t _test is 5 seconds. The maximum density value that can secure the maintenance time T for maintaining the density enhancement value H in the actual inspection is h ₁ .

FIG. 19B shows the case of the injection function inj _{2 in which} the injection rate v is set to a constant value of 3 ml / second equal to the test injection rate v _test and the injection time t ₂ is 10 seconds, which is twice the test injection time t _test. Show. 19 with respect to the test time density curve EHN ₁ of (a), by superimposing the shifted time-density curve obtained which is shifted to 5 seconds for the time axis direction is a test injection time t _test, FIG. 19 ( The estimated time concentration curve enh _{2 of} b) is obtained. In this estimated time concentration curve enh ₂ , the maximum concentration value that can ensure the maintenance time T is h ₂ .

FIG. 19 (c) shows the case of an injection function inj _{6 in which} the injection rate v is set to a constant value 3 ml / second equal to the test injection rate v _test and the injection time t ₆ is 30 seconds, which is six times the test injection time t _test. Show. For the test time concentration curve ehn _{1 in} FIG. 19 (a), this is 5 seconds (1 times the test injection time t _test ), 10 seconds (2 times the test injection time t _test ), 15 seconds (the test injection time). 3 times the t _test), 20 seconds (4 times the test injection time t _test), and 25 seconds (test injection 5x) shifted time-density curve of the five obtained time-axis direction is shifted each time t _test Is superimposed, the estimated time density curve enh ₆ of FIG. 19C is obtained. In this estimated time concentration curve enh ₆ , the maximum concentration value that can ensure the maintenance time T is h ₆ .

Estimated time density curve preparation section 37, similarly, to create an estimated time density curve enh ₃ ~enh ₅ corresponding to three to five times the injection time t ₃ ~t ₅ test injection time. In fact, the estimated time concentration curve ehn _{i (i} is a natural number, injection time is an index representing how many times the test injection time t _test.) Is, with respect to the estimated time concentration curve ehn _i-1 The test time density curve enh _i can be generated by superposing one shift time density curve obtained by shifting the test time density curve enh _i by (i−1) · t _test in the time axis direction.

Reference is again made to FIG.
The maximum density value calculation unit 38 secures the maintenance time T in a plurality of (six in this embodiment) estimated time density curves enh _{1 to} enh ₆ (where enh ₁ is a test time density curve). Maximum density values h ₁ , h ₂ ,..., H ₆ that can be obtained are obtained (step S8). The maintenance time T is a time required for imaging at the actual inspection in the CT scanner 1.

Then, the injection speed calculator 39, the ratio H / h _i (although the concentration-enhancing value H to the maximum density value h ₁ to h ₆ obtained for each estimated time-density curve enh ₁ ~enh _6, i = 1, 2, 3,..., And this is multiplied by the test injection rate v _test to determine the injection rate v _i (= v _test · H / h _i ) corresponding to each injection time t _i (step) S9). The reason why the injection speed v _i is obtained in this way is because it is assumed that the injection function inj (t) and the concentration curve enh (t) are in a linear relationship.

Furthermore, the injection protocol creation unit 40 multiplies the injection speed v _i by the injection time t _i corresponding thereto, thereby giving a contrast agent injection amount a _i (where a ₁ = atest (contrast agent injection at the time of test injection). In this way, six types of injection protocols represented by sets of contrast medium injection amount a _i and injection rate v _i are obtained (step S10), and data representing these six types of injection protocols is stored in the storage unit. This infusion protocol represents a rectangular infusion function corresponding to a single phase infusion where a constant infusion rate v _i is maintained over the infusion time t _i .

FIG. 20 is a diagram for explaining the contents of processing by the injection rate calculation unit 39. FIGS. 20 (a), (b), and (c) correspond to FIGS. 19 (a), (b), and (c), respectively. is doing. By setting the injection rate v _i to be H / h _i times the test injection rate v _test , the required concentration enhancement value H can be secured over the maintenance time T by one-phase injection over each injection time t _i .
The upper area of each of FIGS. 20, infusion rate v _i are represented estimated time-density curve in the case where H / h _i times the test injection velocity v _test and the lower part of each figure, respectively The injection function is shown. Each injection function is stored in the storage unit 8 as described above as an injection protocol represented by a set of contrast medium injection amount and injection speed. The estimated time density curve shown in the upper part of each figure in FIG. 20 is created and stored in the storage unit 8, so that it can be used for controlling the imaging timing by the CT scanner 1 during the actual inspection. In each estimated time density curve, the start time of the maintenance time T is the time to start shooting, and the shooting is ended before the end time of the maintenance time T.

  When six injection protocols are obtained in this way, as shown in FIG. 18, the injection protocol selection unit 41 has one injection protocol that matches the injection conditions (step S11) designated by the operator through the input from the operation input unit 6. Is selected as the injection protocol for the actual inspection (actual injection protocol) (step S12). As injection conditions, for example, the contrast agent injection amount is the minimum, the contrast agent injection speed is the predetermined value or less, the contrast agent injection speed is the minimum, and other appropriate conditions depending on the parameter to be noted. Conditions can be set.

  When the actual injection protocol is determined in this way, the actual injection protocol is transferred from the injection control unit 32 to the control unit 11 of the automatic injector 3. At the same time, the CT scanner 1 is controlled by the scan control unit 31, and imaging for the actual examination is executed by the CT scanner 1 in synchronization with the injection of the contrast agent by the automatic injector 3 (step S13). . The imaging start and end timings at this time are determined based on the estimated time density curve corresponding to the actual injection protocol.

  FIG. 21 is a flowchart for explaining a processing procedure in the case where the estimated time concentration curve creation unit 37 has the configuration of FIG. 13 described above. In FIG. 21, the same reference numerals as those in FIG. 18 are given to the steps where the same processes or operations as those in FIG. 18 are performed. In the process shown in FIG. 21, the following process is performed instead of the process of step S7 in FIG.

That is, the first Fourier transform unit 51 obtains the Fourier transform Inj _test of the test injection function inj _test (step S21). Further, the second Fourier transform unit 52 obtains a Fourier transform Ehh _{test of the} test time concentration curve ehh _test (step S22), and a filter process (high frequency component removal process) is performed by the filter processing unit 54 (step S22). Step S23). Based on the thus determined frequency domain test injection function Inj _test and test time concentration curve Ehh _test , the transfer function G of the subject is determined according to the above equation (13) by the transfer function calculation unit 53 (step S24). Further, an inverse Fourier transform is performed on the transfer function G by the inverse Fourier transform unit 55, whereby a system function g of the subject is obtained (step S25). This system function g represents the impulse response as described above.

Next, a plurality of estimations corresponding to each injection time t _i are performed by superimposing the impulse responses in a time domain over a plurality of different injection times t _i by the action of the overlay processing unit 56 and the shift processing unit 57. time-density curve enh _i are obtained and stored in the storage unit 8 (step S26). The injection time t _{i in} this case need not be a natural number times the test injection time.

The subsequent processing is the same as the processing in steps S8 to S13 described with reference to FIG.
FIG. 22 is a flowchart for explaining a processing procedure in the case where the estimated time concentration curve creation unit 37 has the configuration of FIG. In FIG. 22, the same reference numerals as those in FIG. 21 are given to the steps where the same processes or operations as those in FIG. 21 are performed. In the processing shown in FIG. 22, the following processing is performed instead of the processing in steps S25 and S26 in FIG.

That is, a plurality of injection functions inj _i (representing one-phase injection) in which the injection rate v _i is equal to the test injection rate v _test and the injection time t _i is different from the test injection time t _test are created (step S31). . Then, for each injection function inj _i, by the third Fourier transformer 58, Fourier transform Inj _i injection function inj _i is determined (step S32). Further, using the transfer function G and the injection function Inj _{i in the} frequency domain, the estimated time density curve calculation unit 59 obtains the estimated time density curve Enh _{i in the} frequency domain according to the equation (14) (step S33). . This is by being inverse Fourier transform by the inverse Fourier transform unit 55, is obtained estimated time-density curve enh _i in the time domain, is stored in the storage unit 8 (step S34). By repeated for injecting function inj _i handle multiple of steps S33 and S34, a plurality of estimated time density curve enh _i respectively corresponding to the plurality of injection time t _i is obtained. The injection time t _i need not be a natural number multiple of the test injection time t _test .

The subsequent processing is the same as the processing in steps S8 to S13 described with reference to FIG.
5. Experiment 5-1. Outline of the phantom An experiment was conducted using a phantom that imitates the blood flow of the human abdominal aorta. This phantom has been experimentally confirmed that its time concentration curve is similar to the human time concentration curve.

First, the behavior of the contrast medium injected into the patient, assumed in this phantom, is shown below.
Step 1. A contrast agent is injected into the patient's left elbow vein using an automatic injector. In order to boost the contrast agent that has accumulated in the elbow vein when the injection of the contrast agent is completed, physiological saline is continuously injected by an automatic injector.
Step 2. The injected contrast agent passes through the blood vessel of the left elbow vein and flows into the right atrium of the heart.
Step 3. The contrast agent travels from the right atrium to the right ventricle, pulmonary artery, lung, pulmonary vein, and the left atrium and left ventricle of the heart.
Step 4. The contrast medium flows from the left ventricle of the heart into the abdominal aorta.
Step 5. Contrast media passes through the abdominal aorta. CTA is performed in this part (scanning region), and contrast is performed at a fixed point for a certain period of time to obtain density value fluctuations.
Step 6. After the contrast medium passes through the abdominal aorta, it flows again into the right atrium over the whole body.
Step 7. Return to Step 3.

FIG. 23 shows a schematic diagram of a phantom that mimics the circulation of the contrast agent. here,
v ₀ : Contrast medium injection speed set in the automatic injector (ml / sec)
_Cv : Circulation speed in the phantom (ml / sec)
T _C : Tube volume (ml)
It is.

This phantom models each part of the human body related to changes in the time concentration of the abdominal aorta. In step 1 and 2, the elbow vein from the injection site to the heart is in the tube, in step 3 the cardiopulmonary system from the right atrium to the left ventricle is in the pump and water tank, and in steps 4 and 5 the abdominal aorta is in the scan region P ₁ “Whole body” in FIG. 6 corresponds to each area from the scan region P ₁ to the water tank.

The phantom is filled with water, and water is circulated by a pump at a circulation speed C _v [ml / sec]. In addition, after the contrast agent is injected, the contrast agent in the tube is injected into the phantom by boosting with 20 ml of organized saline at the same speed.
5-2. Phantom Experiment Data In the phantom experiment, Brilliance 40 (40-row multi-slice CT) manufactured by Philips Medical Systems (Cleveland, Ohio, USA) is used as a CT device, and contrast agent is used from an automatic injector (Dual Shot, Nemoto Kyorindo, Tokyo). (Iomeron-350, iodine content 300 mg I / ml, Eisai, Tokyo). Images were taken up to 60 seconds every second after the start of injection, and the concentration value of the abdominal aorta (scan area P _{1 in} FIG. 23) was measured.

In the following calculation, it estimated using the data which set the sampling interval of the measured value for every 1 second to 0.25 second by linear interpolation.
The test time concentration curve for the test injection parameters in Table 1 below is shown in FIG.

  Observation values obtained by actually inspecting the image density by injecting the contrast medium with various injection parameters shown in Table 2 below are indicated by a curve L1 in FIGS. Curves L2 and L3 in each figure show estimated time concentration curves created by the method shown in FIG. However, the curve L2 shows the estimation result using the data of all the observation points up to 60 seconds of the test time concentration curve (see FIG. 24), and the curve L3 is the data up to 20 seconds of the test time concentration curve (see FIG. 24). The estimation result when the density value after 20 seconds is set to “0” is shown.

  From a comparison between the curve L1 and the curves L2 and L3, it can be seen that an estimated time concentration curve that matches the measured value of the time concentration curve can be obtained by the method of the present embodiment.

  Table 3 below shows the results of estimating the injection rate and the like necessary to maintain the 200 HU concentration increase for 10 seconds for the injection time of 10 seconds, 20 seconds and 30 seconds using the test time concentration curve of FIG. Show. From this result, it can be seen that almost the same result is obtained when the data of all observation points of the test time concentration curve is used and when the data of 0 to 20 seconds of the test time concentration curve is used. Therefore, it can be seen that, by extracting and using data of a period in which the concentration change shows a significant change in the test time concentration curve, an appropriate injection parameter can be determined while reducing the calculation load. In addition, since it is sufficient to collect data for a period in which the concentration change shows a significant change in the test time concentration curve, it is possible to set a shorter scan time during the test inspection, thereby reducing the effects of exposure. However, reasonable injection parameters can be determined.

6). Summary As described above, according to this embodiment, the injection rate is set to a constant value equal to the test injection rate, and the injection time is set different from the test injection time for each of a plurality of injection functions based on the test time concentration curve. A time concentration curve is estimated. Then, the maximum concentration value h that can ensure the required concentration maintaining time T in the estimated time concentration curve is obtained, and the injection speed v is obtained based on the ratio of the maximum concentration value h and the required concentration enhancement value H. In this way, it is possible to obtain a plurality of appropriate injection protocols capable of maintaining the concentration enhancement value H over the concentration maintenance time T without performing fitting empirically. Moreover, since this injection protocol is compatible with single-phase injection, it can be used in actual medical practice.
7). Modification In the above-described embodiment, the injection protocol to be used for the actual inspection is automatically selected by setting the conditions for the injection protocol. However, the operator can change from a plurality of injection protocols to the actual inspection. The injection protocol to be used may be manually selected.

In the above-described embodiment, an example in which the present invention is applied to tomography using X-rays has been described. However, the present invention is a system for other imaging methods such as MRI (magnetic resonance imaging) and ultrasonic imaging diagnosis. It can also be applied to.
Further, in the above-described embodiment, an example of a system in which the CT scanner 1, the operation console 2, and the automatic injector 3 cooperate is described. However, the test injection protocol and the computer separated from the CT scanner 1 and the automatic injector 3 A contrast medium injection protocol calculation device that calculates the actual injection protocol based on the test time concentration curve may be configured, or such a computer and an automatic injector are connected to provide a contrast with an injection protocol calculation function. An agent automatic injection device may be configured.

  In addition, various design changes can be made within the scope of matters described in the claims.

It is a block diagram which shows the structural example of the imaging | photography system which concerns on one Embodiment of this invention. It is a conceptual diagram which shows an example of the scanning method by CT scanner. It is a conceptual diagram for demonstrating the back projection method which is one of the methods of estimating the attenuation coefficient distribution from transmission X-ray data (projection data). It is a conceptual diagram for demonstrating a filter correction | amendment back projection method and a convolution back projection method. It is a figure which shows the example of a contrast agent injection function and a time concentration curve. It is a conceptual diagram which shows the view of the system function g showing a human body. It is a figure for demonstrating the deriving process of the transfer function G showing a human body. It is a figure for demonstrating the estimation process of a time concentration curve. It is a figure for demonstrating the estimation process of an ideal injection function. It is a block diagram for demonstrating the functional structure of the data processing part with which the operation console was equipped. It is a figure for demonstrating the process which calculates | requires an estimation time density | concentration curve by the superimposition of a test time density | concentration curve. It is a block diagram for demonstrating the structural example of an estimated time density | concentration curve preparation part. It is a block diagram for demonstrating the other structural example of an estimated time density | concentration curve preparation part. It is a figure which shows a system function (impulse response). It is a figure which shows the example of the estimation time density | concentration curve produced from the impulse response. An example of the estimated time density curve created by the configuration of FIGS. 12 and 13 is shown. It is a block diagram for demonstrating the further another structural example of an estimated time density | concentration curve preparation part. FIG. 13 is a flowchart for explaining a processing procedure for obtaining an injection protocol for a real test when an estimated time concentration curve creation unit has the configuration of FIG. 12. It is a figure for demonstrating the process which calculates | requires the maximum density value which can ensure a required maintenance time from an estimation time density | concentration curve. It is a figure for demonstrating the process which calculates | requires injection | pouring speed | velocity from a required density | concentration enhancement value and the said maximum density | concentration value. It is a flowchart for demonstrating the process sequence in case an estimated time density | concentration curve preparation part has the structure of FIG. It is a flowchart for demonstrating the process sequence in case an estimated time density | concentration curve preparation part has the structure of FIG. It is the schematic which shows the structure of the phantom which imitated the circulation of the contrast agent in a human body. Shows test time concentration curve for test injection parameters in Table 1 It is a figure which shows the example of the time density | concentration curve calculated | required by measurement, and an estimated time density | concentration curve.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 CT scanner 2 Operation console 3 Automatic injector 5 Display apparatus 6 Operation input part 7 Data processing part 8 Memory | storage part 9 Interface part 10 Injection head 11 Control unit 12 Interface part 15 Subject 16 X-ray 17 X-ray tube 18 Detector 21 Test time concentration curve 22 Shift time concentration curve 23 Estimated time concentration curve 31 Scan control unit 32 Injection control unit 33 Image reconstruction unit 34 Injection protocol calculation unit 36 Test time concentration curve generation unit 37 Estimated time concentration curve generation unit 38 Maximum concentration value Calculation unit 39 Injection rate calculation unit 40 Injection protocol creation unit 41 Injection protocol selection unit 45 Shift processing unit 46 Overlay processing unit 51 First Fourier transform unit 52 Second Fourier transform unit 53 Transfer function calculation unit 54 Filter processing unit 55 Inverse Fourier Conversion unit 56 Superposition Processing section 57 shift processing unit 58 third Fourier transform unit 59 estimates the time-density curve calculation unit 60 inverse Fourier transform unit

Claims (13)

A test injection of a contrast medium for a predetermined test injection time at a constant test injection speed for an imaging target;
Imaging the region of interest to be imaged and creating a test time concentration curve representing the temporal change in image concentration associated with a test injection of contrast agent;
Based on the test time concentration curve, the injection rate is set to a constant value equal to the test injection rate, and a time concentration curve is estimated when a predetermined injection time different from the test injection time is set. Creating a curve;
Obtaining a maximum concentration value capable of securing a required concentration maintenance time in the estimated time concentration curve;
Determining an injection rate corresponding to the predetermined injection time so that the required concentration value can be maintained over the concentration maintenance time based on the maximum concentration value and the required concentration value;
Determining a contrast agent injection amount based on the predetermined injection time and a corresponding injection rate, and creating an injection protocol including the contrast agent injection amount and the corresponding injection rate pair. Agent injection protocol determination method. The predetermined injection time is a natural number multiple of the test injection time;
The step of creating the estimated time concentration curve superimposes the test time concentration curve and one or more shift time concentration curves obtained by shifting the test time concentration curve by a natural number times the test injection time on the time axis. The method for determining a contrast medium injection protocol according to claim 1, further comprising the step of creating an estimated time concentration curve corresponding to the predetermined injection time by combining them. Creating the estimated time concentration curve comprises:
Obtaining an impulse response curve which is a time concentration curve corresponding to a contrast agent injection for a predetermined minute unit time based on the test time concentration curve;
2. The method for determining a contrast medium injection protocol according to claim 1, further comprising the step of obtaining the estimated time concentration curve by shifting a plurality of the impulse response curves on the time axis and superimposing a plurality of the impulse response curves according to the predetermined injection time. The step of obtaining the impulse response curve includes:
Fourier transforming a test injection function that represents a time change in injection speed during test injection;
Fourier transforming the test time concentration curve;
Obtaining a transfer function representing a relationship between an injection function representing a change in injection rate over time and a time concentration curve by taking a ratio of each Fourier transform of the test injection function and the test time concentration curve;
The method for determining a contrast medium injection protocol according to claim 3, further comprising: obtaining the impulse response curve by performing inverse Fourier transform on the transfer function. Creating the estimated time concentration curve comprises:
Fourier transforming a test injection function that represents a time change in injection speed during test injection;
Fourier transforming the test time concentration curve;
Obtaining a transfer function representing a relationship between an injection function representing a change in injection rate over time and a time concentration curve by taking a ratio of each Fourier transform of the test injection function and the test time concentration curve;
Fourier transforming a set injection function that is an injection function when the injection rate is a constant value equal to the test injection rate and the injection time is set to the predetermined injection time;
The method for determining a contrast medium injection protocol according to claim 1, further comprising: creating an estimated time concentration curve corresponding to the set injection function based on a Fourier transform of the set injection function and the transfer function. Further comprising the step of performing a filtering process to remove high frequency components with respect to the Fourier transform of the test time concentration curve,
6. The method of determining a contrast agent injection protocol according to claim 4 or 5, wherein the step of determining the transfer function is performed using a Fourier transform of the filtered test time concentration curve. The predetermined injection time includes a plurality of different injection times,
The step of creating the estimated time concentration curve is a step of creating a plurality of estimated time concentration curves respectively corresponding to the plurality of injection times,
The step of obtaining a maximum concentration value capable of securing the concentration maintenance time is a step of obtaining a maximum concentration value capable of securing the concentration maintenance time for each of the plurality of estimated time concentration curves,
The step of determining the injection rate includes a plurality of injection rates respectively corresponding to the plurality of injection times based on a plurality of maximum concentration values and the required concentration values obtained for each of the plurality of estimated time concentration curves. It is a step to seek,
The step of creating the injection protocol is a step of creating a plurality of injection protocols each including a plurality of pairs of the plurality of injection speeds and a plurality of contrast agent injection amounts corresponding to the plurality of injection rates, respectively. The method for determining a contrast medium injection protocol according to any one of the above.   The contrast medium injection protocol determination method according to claim 7, further comprising: extracting an injection protocol whose injection rate is a value within a predetermined range from the plurality of injection protocols.   The method for determining a contrast medium injection protocol according to claim 7 or 8, further comprising: extracting an injection protocol having a lowest injection rate from the plurality of injection protocols.   The method for determining a contrast medium injection protocol according to claim 7, further comprising: extracting an injection protocol having the lowest contrast medium injection amount from the plurality of injection protocols. Contrast agent injection for the actual examination based on the test time concentration curve obtained by imaging the region of interest when the contrast agent is test-injected for a predetermined test injection time at a constant test injection speed for the imaging target A device for computing a protocol,
Based on the test time concentration curve, the injection rate is set to a constant value equal to the test injection rate, and a time concentration curve is estimated when a predetermined injection time different from the test injection time is set. An estimated time concentration curve creating means for creating a curve;
Maximum density value calculating means for obtaining a maximum density value capable of securing a required density maintenance time in the estimated time density curve created by the estimated time density curve creating means;
Based on the maximum concentration value and the required concentration value calculated by the maximum concentration value calculating means, an injection rate that determines an injection rate corresponding to the predetermined injection time so that the required concentration value can be maintained over the concentration maintenance time. Speed calculation means;
A contrast medium injection amount is obtained based on the predetermined injection time and the injection speed calculated by the injection speed calculation means corresponding to the predetermined injection time, and the injection includes a pair of the contrast medium injection amount and the corresponding injection speed. A contrast medium injection protocol computing device, including injection protocol creation means for creating a protocol. A contrast medium injecting means for injecting a contrast medium into an imaging target;
Test injection control means for controlling the contrast medium injection means so that the contrast medium is injected into the imaging object according to a test injection protocol corresponding to the test injection speed and the test injection time;
A contrast medium injection protocol computing device according to claim 11,
A contrast agent injection device including a real injection control unit that controls the contrast agent injection unit so that the contrast agent is injected into the imaging target in accordance with an injection protocol calculated by the contrast agent injection protocol calculation device; The contrast medium injection device according to claim 12,
Photographing means for photographing a photographing object and generating a signal corresponding to the image density;
Test imaging control means for controlling the imaging means to image the region of interest of the imaging object at the time of test injection;
An imaging system including: a real imaging control unit that controls the imaging unit to image the region of interest of the imaging target in synchronization with the contrast medium injection unit controlled by the real injection control unit; .

Images