JP2011005162A - Blood flow dynamic analysis apparatus, magnetic resonance imaging apparatus, blood flow dynamic analysis method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a blood flow dynamic analysis apparatus, a magnetic resonance imaging apparatus and a blood flow dynamic analysis method that can enhance the reliability of a computed perfusion parameter value.SOLUTION: A signal intensity data sequence indicating the time change of signal intensity of magnetic resonance signals is obtained, and the signal intensity data sequence is used to obtain a concentration data sequence indicating the time change of concentration of a contrast medium. Three constants of a fitting function are then computed using the concentration of the contrast medium weighted with a weighting coefficient which is a coefficient defined to reduce the weight of concentration of the contrast medium as it gets far from the peak value time. These constants are used to compute a perfusion parameter for analyzing the blood flow dynamics of a subject.

Description

  The present invention collects a magnetic resonance signal at each time from a subject into which a contrast medium has been injected, and analyzes a blood flow dynamics of the subject based on the collected magnetic resonance signal. The present invention relates to an imaging apparatus, a blood flow dynamic analysis method, and a program.

A method of analyzing a blood flow dynamics of the brain by injecting a contrast medium into a subject is known (see Patent Document 1). When analyzing cerebral blood flow dynamics, for example, a constant of a Γ-variate function is calculated based on a temporal change in the concentration of a contrast agent, and perfusion parameters (for example, average transit time MTT (Mean Transit) are calculated based on the calculated constant. Time), cerebral blood volume CBV (Cerebral Blood Volume), cerebral blood flow CBF (Cerebral
Blood Flow)) is estimated. As a method for estimating the perfusion parameter, there is a first moment method using a first moment of the Γ-variate function. As a first moment method, a method of calculating a constant of a Γ variable function without fitting is known.

Special table 2008-525074

  However, in the method of calculating the constant of the Γ-variate function without performing fitting in the first-moment method, there is a case where the signal-to-noise ratio (S / N ratio) of the magnetic resonance signal collected from the subject is low or the retention of the contrast agent. If it occurs, there is a problem that the reliability of the calculated value of the perfusion parameter is lowered.

  An object of the present invention is to provide a blood flow dynamic analysis device, a magnetic resonance imaging apparatus, a blood flow dynamic analysis method, and a program capable of improving the reliability of the calculated perfusion parameter value in view of the above circumstances. And

The first hemodynamic analysis device of the present invention that solves the above problem is as follows.
A blood flow dynamics analysis device that collects magnetic resonance signals at each time from a subject into which a contrast medium has been injected, and analyzes the blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculating means for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
Fitting means for calculating a constant of a fitting function using a concentration data series in which data representing the concentration of the contrast agent at each time is arranged in time series;
Using the calculated constant of the fitting function, parameter calculating means for calculating a parameter for analyzing the dynamics of the blood flow of the subject;
Have
The fitting means includes
The concentration of the contrast agent is set so that the weight of the concentration of the contrast agent in the peak portion appearing in the density data series is larger than the weight of the concentration of the contrast agent in a portion that is delayed in time from the peak portion. Weighting is performed, and a constant of the fitting function is calculated based on the weighted concentration of the contrast agent.
The first magnetic resonance imaging apparatus of the present invention has the first blood flow dynamic analysis apparatus of the present invention.

The second hemodynamic analysis device of the present invention comprises:
A blood flow dynamic analysis device that collects magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzes the blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculating means for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
Logarithmic processing means for obtaining a logarithmic value of the concentration of the contrast agent at each calculated time;
Fitting means for calculating a constant of a fitting function using a logarithmic value data series in which data representing logarithmic values of the concentration of the contrast agent at each time is arranged in time series;
Using the calculated constant of the fitting function, parameter calculating means for calculating a parameter for analyzing the dynamics of the blood flow of the subject;
Have
The fitting means includes
The weight of the logarithmic value of the concentration of the contrast agent in the peak portion appearing in the logarithmic value data series is larger than the weight of the logarithmic value of the concentration of the contrast agent in a portion delayed in time from the peak portion. The logarithm value of the contrast agent concentration is weighted, and the constant of the fitting function is calculated based on the weighted logarithm value of the contrast agent concentration.
The second magnetic resonance imaging apparatus of the present invention has the second blood flow dynamic analysis apparatus of the present invention.

The first hemodynamic analysis method of the present invention comprises:
A blood flow dynamics analysis method for collecting magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzing blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculating step for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A fitting step of calculating a constant of a fitting function using a concentration data series in which data representing the concentration of the contrast agent at each time is arranged in time series;
A parameter calculating step for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
Have
The fitting step includes
The concentration of the contrast agent is set so that the weight of the concentration of the contrast agent in the peak portion appearing in the density data series is larger than the weight of the concentration of the contrast agent in a portion that is delayed in time from the peak portion. Weighting is performed, and a constant of the fitting function is calculated based on the weighted concentration of the contrast agent.

Moreover, the second blood flow dynamic analysis method of the present invention comprises:
A blood flow analysis method for collecting a magnetic resonance signal at each time from a subject into which a contrast agent has been injected, and analyzing the blood flow dynamics of the subject based on the collected magnetic resonance signal,
A contrast agent concentration calculating step for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A logarithmic processing step for obtaining a logarithmic value of the concentration of the contrast agent at each calculated time;
A fitting step of calculating a constant of a fitting function using a logarithmic value data series in which data representing logarithmic values of the concentration of the contrast agent at each time is arranged in time series;
A parameter calculating step for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
Have
The fitting step includes
The weight of the logarithmic value of the concentration of the contrast agent in the peak portion appearing in the logarithmic value data series is larger than the weight of the logarithmic value of the concentration of the contrast agent in a portion delayed in time from the peak portion. The logarithmic value of the contrast agent concentration is weighted, and the constant of the fitting function is calculated based on the weighted logarithmic value of the contrast agent concentration.

The first program of the present invention is:
A program for collecting magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzing dynamics of blood flow of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculation process for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A fitting process for calculating a constant of a fitting function using a data series in which data representing the concentration of the contrast medium at each time is arranged in time series, and the contrast medium at a peak portion appearing in the data series The concentration of the contrast agent is weighted such that the weight of the concentration of the contrast agent is greater than the weight of the concentration of the contrast agent in the portion delayed in time from the peak portion, and the weighted concentration of the contrast agent is set. A fitting process for calculating a constant of the fitting function based on
A parameter calculation process for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
It is a program for executing.

The second program of the present invention is:
A program for collecting magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzing blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculation process for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A logarithmic conversion process for obtaining a logarithmic value of the concentration of the contrast agent at each calculated time;
A fitting process for calculating a constant of a fitting function using a logarithmic value data series in which data representing logarithmic values of the contrast agent concentration at each time is arranged in a time series, the logarithmic value data series being The concentration of the contrast agent is such that the weight of the logarithmic value of the contrast agent in the peak portion that appears is larger than the weight of the logarithmic value of the concentration of the contrast agent in the portion that is delayed in time from the peak portion. A fitting process for calculating a constant of the fitting function based on a weighted logarithmic value of the contrast agent,
A parameter calculation process for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
It is a program for executing.

In the present invention, the constant of the fitting function is calculated by weighting. Therefore, the reliability of the parameter value for analyzing the calculated dynamics of the blood flow of the subject can be increased.
In the present invention, the constant of the fitting function is a coefficient or an order (for example, Expression (4) or Expression (6)) included in an expression (for example, Expression (4) or Expression (6)) representing the fitting function. It is a number calculated by using a weighted contrast agent concentration (or a logarithmic value of the weighted contrast agent concentration) such as (α, β, γ) included.

1 is a schematic diagram of a magnetic resonance imaging apparatus 1 according to a first embodiment of the present invention. It is a figure which shows the processing flow of the MRI apparatus. 3 is an example of a slice set for a subject 9; It is a conceptual diagram which shows the frame image obtained from slice A1-An. It is a figure showing the time change of the signal strength of the magnetic resonance signal in area | region Ra of slice Ak. It is a figure which shows the peak value time tp in the signal strength data series Dx. It is a figure which shows calculated baseline S0. It is a figure which shows the density | concentration data series Dy showing the time change of the density | concentration C (t) of the contrast agent obtained according to Formula (3). It is a figure which shows (GAMMA) variable function f (t) obtained by substituting the calculated (alpha), (beta), and (gamma) to Formula (4), and the density | concentration data series Dy. It is a figure which shows the MRI apparatus 100 in 2nd Embodiment. It is a figure which shows the processing flow of the MRI apparatus 100 in 2nd Embodiment.

  Hereinafter, although the form for inventing is demonstrated, this invention is not limited to the following forms, A various deformation | transformation is possible.

  The best mode for carrying out the invention will be described below in detail with reference to the drawings.

FIG. 1 is a schematic diagram of a magnetic resonance imaging apparatus 1 according to a first embodiment of the present invention.
A magnetic resonance imaging apparatus (hereinafter referred to as an MRI (Magnetic Resonance Imaging) apparatus) 1 includes a coil assembly 2, a table 3, a contrast medium injection device 4, a reception coil 5, a control device 6, and an input device 7. And a display device 8.

  The coil assembly 2 includes a bore 21 in which the subject 9 is accommodated, a superconducting coil 22, a gradient coil 23, and a transmission coil 24. The superconducting coil 22 applies a static magnetic field B0, the gradient coil 23 applies a gradient pulse, and the transmission coil 24 transmits an RF pulse. Although the superconducting coil 22 is used in the first embodiment, a permanent magnet may be used instead of the superconducting coil 22.

  The table 3 has a cradle 31. The cradle 31 is configured to move in the z direction and the −z direction. The subject 9 is transported to the bore 21 by the cradle 31 moving in the z direction. As the cradle 31 moves in the −z direction, the subject 9 transported to the bore 21 is unloaded from the bore 21.

  The contrast agent injection device 4 injects a contrast agent into the subject 9.

  The receiving coil 5 is attached to the head 9 a of the subject 9. An MR (Magnetic Resonance) signal received by the receiving coil 5 is transmitted to the control device 6.

  The control device 6 includes coil control means 61 to perfusion parameter calculation means 67.

  The coil control unit 61 controls the gradient coil 23 and the transmission coil 24 so that a pulse sequence for imaging the subject 9 is executed in response to an imaging command input from the input device 7 by the operator 10. .

  The signal intensity calculation means 62 calculates the signal intensity of the magnetic resonance signal collected from the subject 9 at each time.

  The time calculation means 63 calculates the time when the signal intensity reaches the peak value in the signal intensity data series Dx (see FIG. 6). Further, the time calculation means 63 calculates the time when the peak portion starts to appear in the signal intensity data series Dx.

  Based on the signal intensity data series Dx, the baseline determining means 64 determines a baseline representing the value of the signal intensity before the contrast agent reaches each region of the subject 9.

  The contrast agent concentration calculating means 65 calculates the concentration of the contrast agent at each time based on the signal intensity of the magnetic resonance signal at each time calculated by the signal intensity calculating means 62.

  The fitting means 66 calculates constants α, β, and γ (see equation (4)) of the fitting function f (t) described later using the density data series Dy (see FIG. 8). Specifically, the contrast medium is weighted so that the weight of the contrast agent concentration in the peak portion PUy appearing in the concentration data series Dy is larger than the weight of the contrast agent concentration in the portion that is delayed in time from the peak portion PUy. The agent concentration is weighted, and constants α, β, and γ of the fitting function f (t) are calculated based on the weighted contrast agent concentration.

  The perfusion parameter calculation unit 67 uses the constant of the fitting function f (t) calculated by the fitting unit 66 to analyze the perfusion parameters (for example, average transit time MTT, brain, etc.). Blood volume CBV, cerebral blood flow volume CBF) are calculated.

  The coil control unit 61 to the perfusion parameter calculation unit 67 are realized by installing a program for executing each unit in the control device 6. However, it may be realized only by hardware without using a program. A combination of the signal intensity calculating means 62, the time calculating means 63, the baseline determining means 64, the contrast medium concentration calculating means 65, the fitting means 66, and the perfusion parameter calculating means 67 included in the control device 6 is the first aspect of the present invention. This corresponds to the blood flow dynamic analysis apparatus according to the first embodiment.

  The input device 7 inputs various commands to the control device 6 in accordance with the operation of the operator 10.

  The display device 8 displays various information and images.

  The MRI apparatus 1 is configured as described above. Next, the operation of the MRI apparatus 1 will be described.

  FIG. 2 is a diagram showing a processing flow of the MRI apparatus 1.

  In step S1, contrast imaging of the head 9a of the subject 9 is performed. The operator 10 operates the input device 7 (see FIG. 1) to set a slice on the subject 9 in order to perform contrast imaging.

  FIG. 3 is an example of a slice set for the subject 9.

  In the subject 9, n slices A1 to An are set. The number of slices is, for example, n = 12. An arbitrary number of slices can be set as necessary.

  After setting the slices A <b> 1 to An, the operator 10 transmits an imaging command for imaging the subject 9 to the coil control means 61 (see FIG. 1) of the MRI apparatus 1. In response to this command, the contrast medium injector 4 injects a contrast medium into the subject 9, and the coil control unit 61 executes a pulse sequence for imaging the head 9 a of the subject 9. The gradient coil 23 and the transmission coil 24 are controlled.

  In the first embodiment, a pulse sequence for obtaining m frame images from each of the slices A1 to An is executed by multi-slice scanning. Therefore, m frame images are obtained for one slice. For example, the number of frame images m = 85. By executing the pulse sequence, frame image data is acquired from the slices A1 to An.

  FIG. 4 is a conceptual diagram showing a frame image obtained from the slices A1 to An.

  FIG. 4A shows frame images in n slices A1 to An set on the head 9a of the subject 9 arranged in time series according to the collection order. In FIG. 4A, the frame images in the slice Ak among the slices A1 to An are indicated by symbols FK1 to FKm. Frame images FK1 to FKm in slice Ak are images collected at times t1 to tm.

  FIG. 4B shows a cross section of the slice Ak and m frame images FK1 to FKm acquired from the slice Ak. Each frame image FK1 to FKm has α × β pixels P1, P2,... Pz. In FIG. 4B, in order to make the pixels P1, P2,... Pz easier to see, the size of the pixels P1, P2,... Pz is increased, but the actual pixels are configured with a smaller size. Yes. Pixels P1, P2,... Pz of the frame images FK1 to FKm correspond to the regions R1, R2,.

  After executing Step S1, the process proceeds to Step S2.

  In step S2, the signal intensity calculating means 62 (see FIG. 1) obtains the time change of the signal intensity of the magnetic resonance signal for each of the regions R1 to Rz of the slices A1 to An of the subject (see FIG. 5).

  FIG. 5 is a diagram illustrating the time change of the signal intensity of the magnetic resonance signal in the region Ra of the slice Ak.

  FIG. 5A shows a cross section of the slice Ak and frame images FK1 to FKm of the slice Ak (see FIG. 4C).

  FIG. 5B shows a signal intensity data series Dx representing a time change of the signal intensity S (t) of the magnetic resonance signal in the region Ra of the slice Ak.

  The horizontal axis is the time t when the frame images FK1 to FKm are acquired from the slice Ak, and the vertical axis is the signal intensity S (t) at the pixel Pa of each frame image FK1 to FKm. The signal strength data series Dx has data DS1 to DSm arranged in time series. Data DS1 to DSm represent signal intensities S (t) in the pixels Pa of the frame images FK1 to FKm, respectively. For example, the data DS1 represents the signal intensity S (t) at the pixel Pa of the frame image FK1, and the data DSg represents the signal intensity S (t) at the pixel Pa of the frame image FKg.

  In the first embodiment, a subject is imaged using a DSC (Dynamic Susceptibility Contrast) method. When the DSC method is used, as shown in FIG. 5B, a convex peak portion PUx appears in the signal intensity data series Dx, where the signal intensity S (t) rapidly decreases and then increases again. .

  FIG. 5 shows the signal intensity data series Dx in the area Ra of the slice Ak, but the signal intensity data series Dx is also obtained for other areas in the slice Ak. Further, the signal intensity data series Dx is similarly obtained for each region of the slice other than the slice Ak. After creating the signal strength data series Dx, the process proceeds to step S3.

  In steps S3 to S8, the same processing is performed on the signal intensity data series Dx obtained for each region R1 to Rn of each slice A1 to An. Therefore, in the following, as a representative, processing performed on the signal intensity data series Dx in the region Ra of the slice Ak will be described.

  In step S3, the time calculation means 63 (see FIG. 1) calculates the peak value time tp when the signal intensity S (t) becomes the peak value (minimum value) in the signal intensity data series Dx.

  FIG. 6 is a diagram illustrating the peak value time tp in the signal intensity data series Dx.

  In the first embodiment, among the data DS1 to DSm of the signal strength data series Dx, the signal strength S21 in the data DS21 is minimized. Therefore, the time calculation means 63 calculates the time t21 in the data DS21 as the peak value time tp. After calculating the peak time tp = t21, the process proceeds to step S4.

  In step S4, the time calculation means 63 calculates a peak part start time ts at which the peak part PUx begins to appear in the signal intensity data series Dx. The peak part start time ts can be calculated using various methods. For example, since the peak portion PUx is convex downward, the time when the signal intensity S (t) becomes higher than the predetermined value S (t) by going back in time from the peak value time tp is obtained. The peak part start time ts can be calculated. In the first embodiment, the time t17 in the data DS17 is calculated as the peak part start time ts. After calculating the peak portion start time ts = t17, the process proceeds to step S5.

  In step S5, the baseline determination means 64 (see FIG. 1) determines a baseline S0 representing the value of the signal intensity before the contrast agent reaches the region Ra of the slice Ak, based on the signal intensity data series Dx. . Baseline S0 is determined as follows.

  When the contrast agent reaches the region Ra of the slice Ak, the signal intensity S (t) of the magnetic resonance signal collected from the region Ra decreases rapidly due to the influence of the contrast agent. Therefore, each data appearing at a time before the peak start time ts is considered to represent the signal intensity S (t) before the contrast agent reaches the region Ra. Therefore, the baseline S0 can be determined by using the signal strength S (t) represented by the data DS1 to DS16. However, since the first data DS1 has a large error, the signal intensity S (t) represented by the data DS1 is excluded from the data for determining the baseline S0, and the baseline S0 is determined using the data DS2 to DS16. .

In the first embodiment, the baseline S0 is determined by calculating the weighted average of the signal strengths S (t) in the data DS2 to DS16. The formula for calculating the baseline S0 is as follows.
S0 = (ΣSi × Wi) / ΣWi (1)
Here, Si: signal intensity S (t) in data DSi
Wi: Weighting coefficient Note that i = 2 to 16.

The weighting coefficient Wi is a coefficient for weighting so that the weight of the signal intensity Si in the data DSi becomes smaller as the data DSi moves away from the peak part start time ts. Such a weight coefficient Wi can be expressed by the following equation, for example.
Wi = k1 / {1 + k2 (t−ts)} (2)
k1 and k2: Arbitrary constants t: Time (t2≤t≤t16)

  By weighting the data DSi using the weighting coefficient Wi represented by the expression (2), the value of the signal strength represented by the baseline S0 can be brought close to the signal strength represented by the data DS17 at the peak start time ts. A highly reliable baseline S0 is obtained.

  FIG. 7 is a diagram showing the calculated baseline S0.

  Based on the baseline S0, the value of the signal intensity before the contrast agent arrives can be determined. After calculating the baseline S0, the process proceeds to step S6.

In step S6, the contrast agent concentration calculating means 65 (see FIG. 1) calculates the contrast agent concentration at times t1 to tm in the region Ra of the slice Ak. The concentration C (t) of the contrast agent at each time t1 to tm is obtained using the following equation.
C (t) = lnS0-lnS (t) (3)
Where lnS0: logarithm of baseline S0
lnS (t): logarithmic value of signal strength S (t)

  FIG. 8 is a diagram showing a concentration data series Dy representing a temporal change in the contrast agent concentration C (t) obtained according to the equation (3).

  The density data series Dy has data DC1 to DCm arranged in time series. Each data DC1 to DCm represents the contrast agent concentration C (t) at each time t1 to tm. After calculating the concentration C (t) of the contrast agent at each time t1 to tm, the process proceeds to step S7.

In step S7, the fitting means 66 (see FIG. 1) calculates the constant of the fitting function using the contrast agent concentration C (t) at each time t1 to tm of the concentration data series Dy (see FIG. 8). In the first embodiment, the Γ variate function f (t) represented by the following equation (4) is used as the fitting function.
f (t) = α (t−ts) ^β · e ^{γ (t−ts)} (4)
However, t ≧ ts
ts: Peak part start time
α, β, and γ: constants determined based on the concentration of contrast agent at t ≧ ts

The Γ variable function f (t) is a fitting function used in the range of t ≧ ts. Therefore, α, β, and γ included in the Γ variable function f (t) are calculated using the value of the contrast agent concentration C (t) in the range of t ≧ ts. However, in the first embodiment, the contrast agent concentration C (t) of t ≧ ts is weighted by the weighting coefficient Wj, and α, β, and γ are calculated using the weighted values. The weighting factor Wj is a factor for weighting the contrast agent concentration C (t) so that the weight of the contrast agent concentration C (t) decreases as the distance from the peak value time tp increases. Such a weight coefficient Wj can be expressed by the following equation, for example.
Wj = k1 / {1 + k2 | t−tp | ^k3 } (5)
k1, k2, and k3: arbitrary constants t: time (t ≧ ts)

  As described above, α, β, and γ are calculated. In the first embodiment, α, β, and γ are calculated using the least square method, but another method other than the least square method may be used. Next, the difference between the Γ variable function f (t) obtained by substituting the calculated α, β, and γ into the equation (4) and the concentration data series Dy will be described with reference to FIG. .

  FIG. 9 is a diagram showing a Γ variable function f (t) obtained by substituting the calculated α, β, and γ into the equation (4), and the concentration data series Dy.

  Referring to FIG. 9, by calculating α, β, and γ using the weighting coefficient Wj expressed by Equation (5), the Γ-variate function f (t) can be calculated in the time range near the peak value time tp. Is a value close to the density data series Dy, but for the time range that is later in time from the peak value time tp, the deviation from the density data series Dy becomes large and a value close to f (t) = 0. Become. However, it is considered that the Γ variable function f (t) shown in FIG. 9 represents the change in the contrast agent concentration with time more accurately than the concentration data series Dy. The reason for this will be described below.

  Referring to the density data series Dy, at the time ts, the contrast agent density C (t) in the region Ra starts to increase gradually. Therefore, it is considered that the bolus of the contrast medium has started to reach the region Ri at time ts. After the time ts, the contrast agent concentration C (t) in the region Ra gradually increases, and the contrast agent concentration C (t) reaches the maximum value at the time tp. Since it flows out from Ra, the concentration C (t) gradually decreases, and it is considered that the concentration of the contrast agent finally becomes C (t) = 0. However, as in the density data series Dy shown in FIG. 9, the density of the contrast agent may have a value that is somewhat larger than C (t) = 0 even after the peak portion PUy. In this case, the density data series Dy is highly reliable as the contrast agent concentration value at the peak portion PUy, but as the contrast agent concentration value at a portion later in time than the peak portion PUy. The reliability is considered low. Therefore, in the first embodiment, the contrast agent concentration C (t) is weighted so that the weight of the contrast agent concentration C (t) decreases as the distance from the peak value time tp increases (see Expression (5)). Α, β, and γ of the Γ-variate function f (t) are calculated using the weighted contrast agent concentration C (t). Therefore, the Γ-variate function f (t) is close to the density data series Dy in the time range in which the peak portion PUy with high reliability as the concentration value of the contrast agent appears, but is longer than the time in which the peak portion PUy appears. At a later time, the value is close to f (t) = 0. Therefore, it is considered that the Γ variable function f (t) shown in FIG. 9 represents the change in the concentration of the contrast agent with time more accurately than the concentration data series Dy.

  After calculating α, β, and γ of the Γ variable function f (t), the process proceeds to step S8, and the perfusion parameter calculation means 67 (see FIG. 1) uses the calculated α, β, and γ to perform various operations. Perfusion parameters (for example, average transit time MTT, cerebral blood volume CBV, cerebral blood flow CBF) are calculated.

  As described above, in the first embodiment, by calculating α, β, and γ using the weighting coefficient Wj shown in Expression (5), the time of the concentration of the contrast agent rather than the concentration data series Dy. A Γ variable function f (t) that accurately represents the change is obtained. Therefore, by calculating various perfusion parameters (for example, average transit time MTT, cerebral blood volume CBV, cerebral blood flow CBF) using α, β, and γ calculated in this way, the calculated perfusion parameters. The reliability of the value of can be increased.

  In the first embodiment, in order to determine the baseline S0, the signal strength S (t) is weighted by using the weighting factor Wi of the equation (2). However, weighting may be performed using a weighting coefficient defined by another expression other than Expression (2).

  Further, in the first embodiment, in order to calculate α, β, and γ included in the Γ variable function f (t), the contrast agent concentration C (t) is expressed by the weighting coefficient Wj in Expression (5). Weighted. However, weighting may be performed using a weighting coefficient defined by another expression other than Expression (5).

  In the first embodiment, since the subject is imaged using the DSC (Dynamic Susceptibility Contrast) method, as shown in FIG. 5B, the signal intensity data series Dx includes the signal intensity S ( A convex peak portion PUx appears when t) decreases rapidly and then increases again. However, a measurement method in which a convex peak portion appears in the signal intensity data series Dx may be used. When a convex peak portion appears in the signal intensity data series Dx, the time when the signal strength reaches the maximum value is the peak value time tp, and the time when the convex peak portion begins to appear is the peak portion. What is necessary is just to calculate (alpha), (beta), and (gamma) as start time ts.

(2) Second Embodiment In the first embodiment, α, β, and γ of the fitting function f (t) of Equation (4) are calculated using C (t) of Equation (3). ing. However, since the fitting function f (t) in Equation (4) is non-linear, it is necessary to perform iterative calculation in order to calculate α, β, and γ. Therefore, the calculation time may be long. Therefore, in the second embodiment, an example in which α, β, and γ can be calculated in a short calculation time will be described.

  First, the principle by which fitting can be performed in a short calculation time in the second embodiment will be described.

In general, when the fitting function is linear, the coefficient of the fitting function can be obtained in a short calculation time. However, the fitting function f (t) used in the first embodiment is a non-linear fitting function. Therefore, the logarithmic transformation of the fitting function f (t) is tried. When f (t) is logarithmically converted, it can be expressed by the following equation (6).
ln {f (t)} = ln (α) + β · ln (t−ts) −γ (t−ts) (6)
However, t ≧ ts

Referring to Equation (6), the right side is expressed linearly. Therefore, α, β, and γ can be calculated in a short time by using ln {f (t)} represented by Expression (6) as a fitting function. However, to obtain the linear form of Equation (6), f (t) must be logarithmically transformed, so that the contrast agent used to calculate α, β, and γ of f (t) The density C (t) (see equation (3)) also needs to be logarithmically converted. When C (t) is logarithmically converted, it is expressed by the following equation (7).
lnC (t) = ln {lnS0-lnS (t)} (7)

  Since both equations (6) and (7) are logarithmically transformed, α, β, and γ of lnf (t) can be calculated by using equation (7). Hereinafter, a second embodiment in which fitting is performed using the equations (6) and (7) will be described.

  FIG. 10 is a diagram illustrating the MRI apparatus 100 according to the second embodiment.

  In description of the MRI apparatus 100 in the second embodiment, differences from the MRI apparatus 1 in the first embodiment will be mainly described.

There are the following two differences between the MRI apparatus 100 of the second embodiment and the MRI apparatus 1 of the first embodiment.
(1) The MRI apparatus 100 of the second embodiment includes a logarithmic conversion unit 651 between the contrast agent concentration calculation unit 65 and the fitting unit 66 as a new component.
(2) The fitting means 66 included in the MRI apparatus 100 of the second embodiment has a different structure from the fitting means 66 included in the MRI apparatus 1 of the first embodiment.

  The signal intensity calculation means 62, time calculation means 63, baseline determination means 64, contrast medium concentration calculation means 65, logarithmic conversion means 651, fitting means 66, and perfusion parameter calculation means 67 included in the control device 6 are combined. The thing corresponds to the blood flow dynamics analysis apparatus in the second embodiment of the present invention.

  The operation of the MRI apparatus 100 in the second embodiment will be described below.

  FIG. 11 is a diagram illustrating a processing flow of the MRI apparatus 100 according to the second embodiment.

  Steps S1 to S6 are the same as the processing flow (see FIG. 2) in the first embodiment, and thus will be described from step S61 below.

  In step S61, the logarithmic conversion means 651 (see FIG. 10) takes the natural logarithm of the contrast agent concentration C (t) (see FIG. 8) in the concentration data series Dy. The natural logarithm of the contrast agent concentration C (t) is expressed by the above-described equation (7). After obtaining lnC (t) represented by Expression (7), the process proceeds to step S7.

  In step S7, the fitting means 66 (see FIG. 10) calculates α, β, and γ of the fitting function ln (f (t)) shown in Expression (6) using lnC (t). In the second embodiment, lnC (t) where t ≧ ts is weighted by the weighting coefficient Wj, and the weighted lnC (t) (the logarithm of the weighted contrast agent concentration) is used to express α, β and γ are calculated. The weighting factor Wj is the same formula as the weighting factor Wj (see formula (5)) used in the first embodiment.

  After calculating α, β, and γ of lnf (t), the process proceeds to step S8, and the perfusion parameter calculation means 67 (see FIG. 10) uses the calculated α, β, and γ to perform various perfusion parameters. (For example, average transit time MTT, cerebral blood volume CBV, cerebral blood flow CBF) are calculated.

  As described above, in the second embodiment, as in the first embodiment, α, β, and γ are calculated using the weighting coefficient Wj shown in Expression (5), and the calculated α, β , And γ are used to calculate various perfusion parameters (eg, mean transit time MTT, cerebral blood volume CBV, cerebral blood flow CBF). Therefore, the reliability of the calculated perfusion parameter value can be increased.

  In the second embodiment, α, β, and γ are calculated by lnf (t) expressed linearly (see Equation (6)). Therefore, α, β, and γ can be calculated without performing iterative processing, and α, β, and γ can be calculated in a short time.

  In the present invention, the fitting function is not limited to f (t) or ln {f (t)}, and other fitting functions can be used within the scope of the present invention.

DESCRIPTION OF SYMBOLS 1,100 Magnetic resonance imaging apparatus 2 Coil assembly 3 Table 4 Contrast agent injection apparatus 5 Reception coil 6 Control apparatus 7 Input apparatus 8 Display apparatus 9 Subject 9a Head 10 Operator 21 Bore 22 Superconducting coil 23 Gradient coil 24 Transmitting coil 31 Cradle 61 Coil control means 62 Signal intensity calculation means 63 Time calculation means 64 Baseline determination means 65 Contrast agent concentration calculation means 66 Fitting means 67 Perfusion parameter calculation means 651 Logarithmic conversion means

Claims (14)

A blood flow dynamics analysis device that collects magnetic resonance signals at each time from a subject into which a contrast medium has been injected, and analyzes the blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculating means for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
Fitting means for calculating a constant of a fitting function using a concentration data series in which data representing the concentration of the contrast agent at each time is arranged in time series;
Using the calculated constant of the fitting function, parameter calculating means for calculating a parameter for analyzing the dynamics of the blood flow of the subject;
Have
The fitting means includes
The concentration of the contrast agent is set so that the weight of the concentration of the contrast agent in the peak portion appearing in the density data series is larger than the weight of the concentration of the contrast agent in a portion that is delayed in time from the peak portion. A blood flow dynamic analysis device that performs weighting and calculates a constant of the fitting function based on the weighted concentration of the contrast agent. Signal intensity calculating means for calculating the signal intensity of the magnetic resonance signal at each time collected from a predetermined region of the subject;
Time calculating means for calculating a first time when the signal intensity reaches a peak value in a signal intensity data series in which data representing the signal intensity at each time is arranged in time series;
Have
The weighting means is
The contrast agent concentration is weighted such that the contrast agent concentration weight at the first time is greater than the contrast agent concentration weight at a second time that is later than the first time. The blood flow dynamic analysis device according to claim 1. The weighting means is
The blood according to claim 2, wherein the concentration of the contrast medium is weighted so that the weight of the concentration of the contrast medium at the second time becomes smaller as the second time moves away from the first time. Flow analysis device.   Based on the signal intensity of the magnetic resonance signal at each time calculated by the signal intensity calculation means, a baseline representing the value of the signal intensity before the contrast agent reaches the predetermined region of the subject is determined. The blood flow dynamic analysis device according to claim 2 or 3, further comprising a baseline determination unit. The time calculating means includes
Calculating a third time when a peak portion begins to appear in the signal strength data series;
The baseline determination means includes
The signal strength is weighted such that the weight of the signal strength at the third time is greater than the weight of the signal strength at the fourth time earlier than the third time, and the weighted signal strength The blood flow dynamic analysis apparatus according to claim 4, wherein the baseline is calculated using a computer. The baseline determination means includes
6. The blood flow dynamics according to claim 5, wherein the concentration of the contrast agent is weighted so that the weight of the signal intensity at the fourth time becomes smaller as the fourth time moves away from the third time. Analysis device. The blood flow dynamic analysis device according to claim 5 or 6, wherein the fitting function is represented by the following equation.
f (t) = α (t−ts) ^β · e ^{γ (t−ts)}
However, t ≧ ts
ts: third time
α, β, and γ: constants determined based on the concentration of contrast agent at t ≧ ts   A magnetic resonance imaging apparatus comprising the blood flow dynamic analysis apparatus according to claim 1. A blood flow dynamic analysis device that collects magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzes the blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculating means for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
Logarithmic conversion means for obtaining a logarithmic value of the concentration of the contrast agent at each calculated time;
Fitting means for calculating a constant of a fitting function using a logarithmic value data series in which data representing logarithmic values of the concentration of the contrast agent at each time is arranged in time series;
Using the calculated constant of the fitting function, parameter calculating means for calculating a parameter for analyzing the dynamics of the blood flow of the subject;
Have
The fitting means includes
The weight of the logarithmic value of the concentration of the contrast agent in the peak portion appearing in the logarithmic value data series is larger than the weight of the logarithmic value of the concentration of the contrast agent in a portion delayed in time from the peak portion. A blood flow kinetic analysis apparatus that weights a logarithmic value of the contrast agent concentration and calculates a constant of the fitting function based on the weighted logarithm value of the contrast agent. The hemodynamic analysis device according to claim 9, wherein the fitting function is represented by the following equation.
ln {f (t)} = ln (α) + β · ln (t−ts) −γ (t−ts)
However, t ≧ ts
ts: Peak part start time
α, β, and γ: constants determined based on the logarithmic value of the contrast agent concentration at t ≧ ts A blood flow dynamics analysis method for collecting magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzing blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculating step for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A fitting step of calculating a constant of a fitting function using a data series in which data representing the concentration of the contrast agent at each time is arranged in time series;
A parameter calculating step for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
Have
The fitting step includes
The contrast agent concentration is weighted such that the contrast agent concentration weight at the peak portion appearing in the data series is greater than the contrast agent concentration weight at a portion delayed in time from the peak portion. And a blood flow dynamic analysis method for calculating a constant of the fitting function based on the weighted concentration of the contrast agent. A blood flow analysis method for collecting a magnetic resonance signal at each time from a subject into which a contrast agent has been injected, and analyzing the blood flow dynamics of the subject based on the collected magnetic resonance signal,
A contrast agent concentration calculating step for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A logarithmic conversion step for obtaining a logarithmic value of the concentration of the contrast agent at each calculated time;
A fitting step of calculating a constant of a fitting function using a logarithmic value data series in which data representing logarithmic values of the concentration of the contrast agent at each time is arranged in time series;
A parameter calculating step for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
Have
The fitting step includes
The weight of the logarithmic value of the concentration of the contrast agent in the peak portion appearing in the logarithmic value data series is larger than the weight of the logarithmic value of the concentration of the contrast agent in a portion delayed in time from the peak portion. A blood flow dynamics analysis method for weighting a logarithmic value of the contrast agent concentration and calculating a constant of the fitting function based on the weighted logarithmic value of the contrast agent. A program for collecting magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzing dynamics of blood flow of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculation process for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A fitting process for calculating a constant of a fitting function using a data series in which data representing the concentration of the contrast medium at each time is arranged in time series, and the contrast medium at a peak portion appearing in the data series The concentration of the contrast agent is weighted such that the weight of the concentration of the contrast agent is greater than the weight of the concentration of the contrast agent in the portion delayed in time from the peak portion, and the weighted concentration of the contrast agent is set. A fitting process for calculating a constant of the fitting function based on
A parameter calculation process for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
A program for running. A program for collecting magnetic resonance signals at each time from a subject into which a contrast agent has been injected, and analyzing blood flow dynamics of the subject based on the collected magnetic resonance signals,
A contrast agent concentration calculation process for calculating the concentration of the contrast agent at each time of the predetermined region based on the signal intensity of the magnetic resonance signal at each time collected from the predetermined region of the subject;
A logarithmic conversion process for obtaining a logarithmic value of the concentration of the contrast agent at each calculated time;
A fitting process for calculating a constant of a fitting function using a logarithmic value data series in which data representing logarithmic values of the contrast agent concentration at each time is arranged in a time series, the logarithmic value data series being The concentration of the contrast agent is such that the weight of the logarithmic value of the contrast agent in the peak portion that appears is larger than the weight of the logarithmic value of the concentration of the contrast agent in the portion that is delayed in time from the peak portion. A fitting process for calculating a constant of the fitting function based on a weighted logarithmic value of the contrast agent,
A parameter calculation process for calculating a parameter for analyzing the dynamics of the blood flow of the subject using the calculated constant of the fitting function;
A program for running.

Images