WO2016194899A1 - Magnetic resonance imaging device and high-frequency magnetic field shimming method - Google Patents

Abstract

Provided is an RF shimming technique whereby maximum effect is obtained irrespective of purpose even when measurement noise is included in an MRI device which uses a transmission coil having a plurality of channels. In the present invention, RF shimming is performed using an optimization index whereby the optimum RF parameters can be calculated even when measurement noise is included in the RF shimming. The optimization index is obtained by weighting the B₁ value in accordance with at least one of the position of an imaging region and the size of the initial B₁ value. The weighting is performed by multiplying the B₁ value by a weighting function, for example.

Description

Magnetic resonance imaging apparatus and high-frequency magnetic field shimming method

The present invention relates to a magnetic resonance imaging (MRI) technique, and more particularly to a high-frequency magnetic field irradiation technique for generating a rotating magnetic field that induces a magnetic resonance phenomenon.

The MRI apparatus is a medical image diagnostic apparatus that causes magnetic resonance to occur in nuclei in an arbitrary cross section that crosses the examination target, and obtains a tomographic image in the cross section from the generated magnetic resonance signal. A radio wave (Radio Frequency wave, hereinafter referred to as RF), which is a type of electromagnetic wave, is transmitted to the inspection object to excite the spins of the nuclei in the inspection object, and then receives a nuclear magnetic resonance signal generated by the nuclear spins. The inspection object is imaged. RF transmission to the inspection object is performed by the RF transmission coil, and reception of the nuclear magnetic resonance signal from the inspection object is performed by the RF reception coil.

In recent years, with the aim of improving the SNR (Signal to Noise Ratio) of images, the static magnetic field strength tends to increase, and the spread of high magnetic field MRI devices (super high magnetic field MRI devices) with a static magnetic field strength of 3 T (Tesla) or higher. Has begun. However, as the static magnetic field strength increases, the SNR improves, but uneven brightness tends to occur in the captured image. This is because the RF frequency used for inducing the magnetic resonance phenomenon increases with the increase in the magnetic field, and the irradiation RF distribution and the magnetic resonance phenomenon generated by the RF in relation to the size of the inspection object. This is because the spatial distribution of the rotating magnetic field that induces non-uniformity. This is called non-uniformity of the transmission sensitivity distribution (high-frequency magnetic field distribution, B ₁ distribution).

If the B ₁ distribution is non-uniform, uneven brightness occurs. If this non-uniformity is large, the fat signal cannot be sufficiently suppressed, and the effect of removing the fat signal may be insufficient.

There is a technique called “RF shimming” as an RF irradiation method for reducing nonuniformity of the B ₁ distribution. This is a technique for reducing B ₁ nonuniformity in an imaging region by using a transmission coil having a plurality of channels and controlling the phase and amplitude (hereinafter referred to as an RF parameter) of an RF pulse applied to each channel. (For example, refer to Patent Document 1). There has also been proposed a method of calculating an RF parameter that makes the B ₁ distribution uniform with higher accuracy by changing the gradient magnetic field waveform (see, for example, Patent Document 2).

As a method for calculating the RF parameter for reduction of the B ₁ uneven, define an index of B ₁ uniformity, there is a method for solving an optimization problem for minimizing the value. As the B ₁ uniformity index at that time, the standard deviation of the B ₁ value (see, for example, Patent Document 3), the B ₁ maximum value, the B ₁ minimum value, and the like are used.

US Pat. No. 7,078,901 US Patent Application Publication No. 2003/0214294 Special table 2012-502683 gazette

The purpose of RF shimming is not only to reduce the non-uniformity of the B ₁ distribution, but also to suppress SAR and suppress fat signals. Among these, when the fat signal is suppressed, the difference between the maximum value and the minimum value of the B ₁ value is often used as an index. If the B ₁ non-uniformity is large and deviates from the reference B ₁ value, fat suppression cannot be performed. Therefore, the RF parameter is determined so as to make the index as small as possible.

The B ₁ value is easily affected by noise components (measurement noise) generated during imaging. When the standard deviation is used as an index when determining the RF parameter, the influence of the measurement noise is small if the number of points where the measurement noise enters is sufficiently small relative to the number of measurement points of the B ₁ value. However, as the time of the fat signal suppression, when using the difference between the maximum and minimum values of B ₁ value in the index under the influence of the measurement noise, the extremely small value or large value is mixed even one, It is greatly affected when RF parameters are determined, and optimal parameters cannot be obtained, and fat cannot be sufficiently suppressed even in an image.

The present invention has been made in view of the above circumstances, and in an MRI apparatus using a transmission coil having a plurality of channels, even if measurement noise is included, RF shimming that achieves the maximum effect regardless of the purpose. The purpose is to provide technology.

The present invention performs RF shimming using an optimization index capable of calculating an optimum RF parameter even when measurement noise is included in RF shimming. Optimization index, the B ₁ value, obtained by performing weighting according to at least one of the magnitude of the position and initial B ₁ value of the imaging region. Weighting, for example, be done by multiplying the weighting function to the B ₁ value.

According to the present invention, in an MRI apparatus using a transmission coil having a plurality of channels, even if measurement noise is included, RF shimming that achieves the maximum effect can be realized regardless of the purpose.

It is a block diagram of the MRI apparatus of embodiment of this invention. It is explanatory drawing for demonstrating the transmission coil of embodiment of this invention. It is a functional block diagram of the computer of the embodiment of the present invention. (A) is, embodiments of the present invention, is an explanatory diagram for explaining the B ₁ value used when creating optimization metrics, (B) it is for illustrating the weighting function of the embodiment of the present invention It is explanatory drawing. It is a flowchart of the imaging process of embodiment of this invention. It is explanatory drawing for demonstrating the setting area | region in the human body model used for the electromagnetic simulation of embodiment of this invention. (A) to (F), when there is no measurement noise, no RF shimming, RF shimming is performed using the standard deviation index U _SD , and RF shimming is performed using the maximum / minimum index U _NEMA In this case, each B ₁ distribution and a binarized map are explanatory diagrams for explaining the case. (A) ~ (C) is a state measurement noise is not, no RF shimming, in the case of performing the RF shimming using standard deviation index U _SD, and were RF shimming using a maximum minimum index U _NEMA In each case, a histogram of B ₁ values in the human body model. (A) to (F), when there is measurement noise, no RF shimming, RF shimming is performed using the standard deviation index U _SD , and RF shimming is performed using the maximum / minimum index U _NEMA In this case, each B ₁ distribution and a binarized map are explanatory diagrams for explaining the case. (A) ~ (C), in a state in which measurement noise is present, no RF shimming, in the case of performing the RF shimming using standard deviation index U _SD, and were RF shimming using a maximum minimum index U _NEMA In each case, a histogram of B ₁ values in the human body model. (A) ~ (F) is a state measurement noise is not, in the case of performing the RF shimming using standard deviation index U _SD, in the presence of measurement noise, performing RF shimming using a maximum minimum index U _NEMA When the RF shimming is performed using the optimization index U _WSD of the present embodiment in a state where there is measurement noise, the B ₁ distribution and the binarization map will be described. (A) ~ (C) is a state measurement noise is not, in the case of performing the RF shimming using standard deviation index U _SD, in the presence of measurement noise, performing RF shimming using a maximum minimum index U _NEMA In the case where RF shimming is performed using the optimization index U _WSD of the present embodiment in the presence of measurement noise, each is a histogram of B ₁ values in the human body model. (A) to (E) are explanatory diagrams for explaining modifications of the weighting function according to the embodiment of the present invention.

Hereinafter, an example of an embodiment to which the present invention is applied will be described with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiments unless otherwise specified, and repetitive description thereof is omitted. In addition, this invention is not limited by this embodiment.

[Configuration of MRI system]
First, the overall configuration of the MRI apparatus of this embodiment will be described. FIG. 1 is a block diagram of the MRI apparatus 100 of the present embodiment.

As shown in the figure, the MRI apparatus 100 of the present embodiment includes a magnet 101 that generates a static magnetic field, a gradient coil 102 that generates a gradient magnetic field, a shim coil 112 that adjusts the static magnetic field uniformity, a sequencer 104, An RF transmission coil (transmission coil) 114 that irradiates (transmits) a high-frequency magnetic field (RF); an RF reception coil (reception coil) 115 that detects (receives) a nuclear magnetic resonance signal generated from the subject 103; A table 107 on which the subject 103 is placed, a gradient magnetic field power source 105, a high-frequency magnetic field generator 106, a receiver 108, a shim power source 113, and a computer 109 that controls each part of the MRI apparatus 100 and realizes imaging. .

The gradient magnetic field coil 102 and the shim coil 112 are connected to the gradient magnetic field power source 105 and the shim power source 113, respectively. The transmission coil 114 and the reception coil 115 are connected to the high-frequency magnetic field generator 106 and the receiver 108, respectively.

The sequencer 104 sends a command to the gradient magnetic field power source 105, the shim power source 113, and the high frequency magnetic field generator 106 according to an instruction from the computer 109 to generate a gradient magnetic field and RF, respectively. RF is irradiated (transmitted) to the subject 103 through the transmission coil 114.

The nuclear magnetic resonance signal generated from the subject 103 by irradiating (transmitting) RF is detected (received) by the receiving coil 115 and detected by the receiver 108. A magnetic resonance frequency used as a reference for detection by the receiver 108 is set by the computer 109 via the sequencer 104.

The detected signal is sent to the computer 109 through an A / D conversion circuit, where signal processing such as image reconstruction is performed. The result is displayed on the display device 110 connected to the computer 109. The detected signals and measurement conditions are stored in the storage device 111 connected to the computer 109 as necessary.

The magnet 101, shim coil 112, and shim power supply 113 constitute a static magnetic field forming unit that forms a static magnetic field space. The gradient magnetic field coil 102 and the gradient magnetic field power source 105 constitute a gradient magnetic field application unit that applies a gradient magnetic field to the static magnetic field space. The transmission coil 114 and the high-frequency magnetic field generator 106 constitute a high-frequency magnetic field transmission unit that irradiates (transmits) a high-frequency magnetic field (RF) to the subject 103 arranged in the static magnetic field. The receiving coil 115 and the receiver 108 constitute a signal receiving unit that detects (receives) a nuclear magnetic resonance signal generated from the subject 103.

The transmission coil 114 of the present embodiment is a multi-channel coil that includes a plurality of channels that independently transmit a high-frequency magnetic field (RF) to the subject 103. FIG. 2 shows an example of the transmission coil 114 of the present embodiment. Here, a case where the transmission coil 114 is a four-channel (4ch) coil including four channels (114a, 114b, 114c, 114d) is illustrated.

The amplitude and phase of RF transmitted from each channel (114a, 114b, 114c, 114d) are individually set by the computer 109. The high-frequency magnetic field generator 106 according to the present embodiment is independent of each channel via feeding points (117a, 117b, 117c, 117d) included in each channel (114a, 114b, 114c, 114d) in accordance with control from the computer 109. An RF waveform (RF pulse) is transmitted. In this figure, reference numeral 116 denotes an RF shield.

As described above, the computer 109 according to the present embodiment controls each unit of the MRI apparatus 100 to realize imaging. In the present embodiment, a static magnetic field shimming process for adjusting the uniformity of the static magnetic field in the imaging space and an RF shimming process for adjusting the uniformity of the B ₁ distribution in the region of interest according to the purpose are further performed.

[Computer function blocks]
In order to realize these, the computer 109 according to the present embodiment collects image data according to the imaging condition setting unit 210 that sets the imaging conditions and the imaging conditions set by the imaging condition setting unit 210, as shown in FIG. An imaging unit 220 that performs imaging. In addition, the imaging condition setting unit 210 includes an imaging position setting unit 211 that sets an imaging position, a static magnetic field shimming unit 212 that performs static magnetic field shimming processing, and an RF shimming unit 213 that performs RF shimming processing.

Each function realized by the computer 109 is realized by a CPU included in the computer 109 loading a program stored in advance in the storage device 111 to the memory and executing the program.

It should be noted that all or some of the functions may be realized by hardware such as ASIC (Application Specific Integrated Circuit) or FPGA (field-programmable gate array). Various data used for processing of each function and various data generated during the processing are stored in the storage device 111.

Hereinafter, details of each unit of the imaging condition setting unit 210 of the present embodiment will be described.

[Imaging position setting section]
The imaging position setting unit 211 sets an imaging position (imaging cross section). The imaging section is set using a positioning image obtained by performing a scout scan or the like before performing the main imaging. For example, on the positioning image displayed on the display device 110, designation by the user is accepted, and the designated position is set as an imaging section. As an imaging cross section, a predetermined position may be automatically set for each part, using a feature point on the positioning image as a clue. Note that the region of the subject 103 on the imaging section is referred to as an imaging region.

[Static magnetic field shimming section]
The static magnetic field shimming unit 212 measures the static magnetic field distribution and performs adjustment so that the static magnetic field is as uniform as possible. The adjustment is performed by operating the shim coil 112 via the shim power supply 113. If it is not necessary to adjust the uniformity of the static magnetic field, the static magnetic field shimming process may not be performed. Further, the static magnetic field shimming unit 212, the shim power source 113, and the shim coil 112 may not be provided.

[RF shimming section]
The RF shimming unit 213 according to the present embodiment uses the parameters of the high frequency magnetic field (RF) transmitted from each channel of the transmission coil 114 so as to correct the nonuniformity of the B ₁ distribution that is the high frequency magnetic field distribution in the region of interest (ROI). A certain high frequency magnetic field parameter (RF parameter) is determined. Here, at least one of the amplitude and phase of the RF transmitted from each channel of the transmission coil 114 is determined so as to reduce the B ₁ non-uniformity in the ROI according to the purpose. Note that the RF amplitude and phase are collectively referred to as an RF parameter when it is not necessary to distinguish between them.

In the present embodiment, the RF shimming unit 213 determines an optimal RF parameter using an optimization index prepared in advance. In the present embodiment, this index is created using the B ₁ value. Then, the optimization index is weighted according to the magnitude of the B ₁ value is generated by performing the B ₁ value. This weighting is performed by a weighting function.

For this reason, the RF shimming unit 213 of the present embodiment includes an index creating unit 232 that creates an optimization index used for RF shimming, and an RF parameter determination unit 233 that determines an RF parameter using the determined optimization index. Prepare.

[Indicator creation section]
In the present embodiment, as an optimization index used for RF shimming, an index that optimizes the B ₁ distribution in the ROI and maximizes the fat suppression effect even when there is measurement noise is used. As described above, the index creating unit 232 according to the present embodiment creates an optimization index that realizes this using the value of the B ₁ distribution.

[Conventional indicators]
Prior to the description of the optimization index of the present embodiment, the conventional index used for RF shimming will be described below.

In order to optimize the B ₁ distribution in the ROI, the variance of the B ₁ distribution in the ROI may be minimized. Therefore, as an index to achieve this, an indicator using the standard deviation of the B ₁ distribution (hereinafter, referred to as the standard deviation indicator) is U _SD.

The standard deviation index _USD is expressed by the following formula (1).

R is a spatial coordinate, B ₁ (r) is a B ₁ distribution, σ (B ₁ (r)) is a standard deviation of B ₁ value, and m (B ₁ (r)) is an average value of B _1. , Respectively. That is, the standard deviation index U _SD is the standard deviation which is normalized, as the standard deviation index U _SD is small, variation in the values is small, indicating a uniform.

In order to maximize the fat suppression effect, the difference between the maximum value and the minimum value of the B ₁ value in the ROI may be minimized. As an index for realizing this, there is an index that uses the maximum value and the minimum value of the B ₁ value of the B ₁ distribution (hereinafter, maximum and minimum index) U _NEMA .

The index (maximum / minimum index) U _NEMA using the B ₁ maximum value and the B ₁ minimum value is expressed by the following equation (2).

Note that max (B ₁ ) indicates the B ₁ maximum value, and min (B ₁ ) indicates the B ₁ minimum value. That is, the maximum / minimum index U _NEMA is a standardized difference between the B ₁ maximum value and the B ₁ minimum value. The smaller the maximum / minimum index U _NEMA , the smaller the value difference and the more uniform.

However, since the maximum / minimum index U _NEMA uses two specific points of the maximum value and the minimum value, for example, if an extremely small value is mixed under the influence of measurement noise, the accuracy as the index is lowered.

[Indicator of this embodiment]
Therefore, in the present embodiment, the B ₁ distribution in the ROI is optimized, and even if there is measurement noise, the optimization index that maximizes the fat suppression effect is a value within a predetermined range including the minimum value. And an index using a predetermined range including the maximum value.

That is, the index creation unit 232 of the present embodiment uses the B ₁ distribution, and uses a B ₁ value that is smaller than a predetermined threshold (first threshold) and a B ₁ value that is larger than a predetermined threshold (second threshold). Create optimization metrics. The first threshold value is a value that is not less than the minimum value and smaller than the second threshold value, and the second threshold value is not more than the maximum value and larger than the first threshold value.

The B ₁ value used in the optimization index of this embodiment will be described with reference to FIG. In FIG. 4A, 300 is a histogram showing the distribution of B ₁ values. The magnitude of the B ₁ value is represented on the horizontal axis, and the frequency (frequency) of each B ₁ value is represented on the vertical axis.

Index creation unit 232 of the present embodiment uses and B ₁ value in the first threshold value 321 is smaller than range 331, a and B ₁ value in the second threshold 322 is larger than range 332, to create an optimized index.

Such an optimization index can be calculated by weighting the B ₁ value of the B ₁ distribution according to the size of the B ₁ value. Therefore, the index creation unit 232 according to the present embodiment creates an optimization index by determining the weighting function w (B ₁ ) 602 for performing weighting. The weighting function is a function having the B ₁ value as a variable.

An example of the weighting function w (B ₁ ) 310 for extracting the B ₁ value in the range 331 smaller than the first threshold 321 and the B ₁ value in the range 332 larger than the second threshold 322 as described above is shown in FIG. Shown in B). Weighting function w (B ₁₎ 310 shown in the figure, a first threshold value 321 is smaller than B ₁ value determined in advance, the greater first threshold 321, a second threshold value 322 is greater than B ₁ a predetermined It has a shape to extract values. By multiplying the B ₁ distribution by this weighting function w (B ₁ ) 310, only the B ₁ value in the above range is extracted.

The first threshold 321 and the second threshold 322 are, for example, predetermined fixed values and are held in the storage device 111.

The first threshold value 321 and the second threshold value 322 are also calculated from the B ₁ value. For example, the first threshold value 321 is a threshold value for extracting the lower p% value of the B ₁ value, and the second threshold value 322 is a threshold for extracting the upper q% value of the B ₁ value (p and q are Predetermined positive real number). These p and q are determined in advance and held in the storage device 111.

Note that the first threshold 321 and the second threshold 322 are not limited to this. For example, an average value (m) and a standard deviation (σ) of the B ₁ value may be calculated and defined using these values. For example, the first threshold 321 is m−rσ, and the second threshold 322 is m + rσ.

The index creating unit 232 according to the present embodiment determines the weighting function w (B ₁ ) 310 using the first threshold value 321 and the second threshold value 322 obtained by the above-described method, and then the weighting function w (B ₁ ). 310 is multiplied (masked) by the initial B ₁ value to create an optimization index. Optimization metrics U _WSD, for example, as shown in the following equation (3), although the weighting function w (B ₁₎ 310 obtained by multiplying the B ₁ value (position each of B ₁ value (B ₁ (r)) The standard deviation is obtained by dividing by the average value of B ₁ values.

The optimization index U _WSD shown in Expression (3) is an index created using a relatively small B ₁ value and a relatively large B ₁ value. Compared to the fact that the conventional maximum and minimum index U _NEMA was created using two values, the maximum and minimum values of B ₁ , the optimization index U _WSD is the B ₁ value used for index generation. The number increases. That is, the B ₁ value used for index creation increases. As a result, the optimization index U _WSD of the present embodiment is less susceptible to measurement noise than the conventional maximum / minimum index U _NEMA .

Note that the index creation unit 232 of the present embodiment creates the optimization index U _WSD before the RF parameter determination processing by the RF parameter determination unit 233.

Note that the weighting function w (B ₁ ) 310 may be stored in the storage device 111 in advance, and the optimization index U _WSD may be generated using the weighting function w (B ₁ ) 310. Further, the weighting function w (B ₁ ) 310 and the optimization index U _WSD using the weighting function w (B ₁ ) 310 may be stored in the storage device 111 in advance. In this case, the index creation unit 232 may not be provided. An RF parameter determination unit 233 described later determines an RF parameter using an optimization index U _WSD stored in advance in the storage device 111.

[RF parameter determination unit]
The RF parameter determination unit 233 uses the optimization index U _WSD created by the index creation unit 232 to determine an RF parameter that realizes a desired B ₁ distribution.

In the present embodiment, the RF parameter determination unit 233 determines each RF parameter so as to minimize the optimization index U _WSD . That is, the RF parameter determination unit 233 obtains an RF parameter as a solution for minimizing the optimization index U _WSD created by the index creation unit 232. Computation of solutions covers various optimization problems such as steepest descent, gradient, Newton, least squares, conjugate gradient, linear programming, nonlinear programming, amplitude and phase values. A method of calculating an optimal solution by changing the value can be used. Further, the optimization index U _WSD may be approximated by a polynomial by fitting or the like to obtain the minimum value of the approximated function.

The RF shimming unit 213 replaces the RF parameter set as the imaging condition with the RF parameter determined by the RF parameter determination unit 233. Then, the imaging unit 220 performs imaging using the RF parameter determined by the RF shimming unit 213.

[Flow of imaging processing]
Hereinafter, the flow of the imaging process including the RF shimming process of the present embodiment by each function of the computer 109 of the present embodiment will be described. FIG. 5 is a processing flow of the imaging process of the present embodiment. The imaging process of this embodiment is started by an instruction from the user. Here, the static magnetic field shimming process is omitted.

First, the imaging condition setting unit 210 receives an input of imaging conditions including patient information, imaging region, imaging purpose, imaging parameters, and the like from the user (step S1101). Next, the imaging position setting unit 211 performs a scout scan and sets the imaging position (step S1102).

Next, the RF shimming unit 213 performs RF shimming and determines an RF parameter (step S1103). The imaging condition setting unit 210 sets the RF parameters determined by the RF shimming unit 213 as imaging amplitudes together with other imaging parameters as the amplitude and phase of the RF transmitted to each channel used for imaging (step S1104).

Then, the imaging unit 220 performs imaging in accordance with the imaging conditions set by the RF parameter determination unit 233 (step S1105), and ends the process.

<Simulation results>
The homogenization of the B ₁ distribution and the fat suppression effect by the two conventional indexes (standard deviation index U _SD and maximum / minimum index U _NEMA ) and the index U _WSD of this embodiment will be described using electromagnetic field analysis simulation. . Hereinafter, in the description of the electromagnetic simulation results, RF parameters are determined so as to minimize a predetermined optimization index, RF shimming using the index, or RF shimming using the index, etc. Call it.

[Human body model]
In this simulation, the B ₁ distribution generated in the living body was calculated using a numerical human body model (Hugo model), and the homogenization and fat suppression effects were confirmed. FIG. 6 shows a part 401 and the ROI 402 of the numerical human body model used at this time. Here, the B ₁ distribution in the ROI 402 was optimized with a SAG (sagittal) cross section including the lumbar vertebra region 403.

[RF irradiation method]
In this simulation, for example, the four-channel transmission coil 114 shown in FIG. 2 is modeled, and the numerical human body model 401 is placed therein and irradiated with RF. The frequency of the irradiated RF was set to 128 MHz assuming a 3T MRI apparatus.

Moreover, RF (B_ch1, B_ch2, B_ch3, B_ch4) of the sine waveform shown in the following formula (4) is provided at the feeding point (117a, 117b, 117c, 117d) of each channel (114a, 114b, 114c, 114d). Was fed.

A1 and φ1 are the amplitude and phase of the sine waveform voltage supplied to the feeding point 117a of the channel 114a, respectively. A2 and φ2 are the same amplitude and phase supplied to the feeding point 117b of the channel 114b, respectively. A3 and φ3 Are the same amplitude and phase supplied to the feeding point 117c of the channel 114c, and A4 and φ4 are the amplitude and phase supplied to the feeding point 117d of the channel 114d, respectively.

Further, QD (Quadrature Drive) irradiation, which is a standard RF irradiation method, was used for irradiation without RF shimming. That is, A1, A2, A3, and A4 are all 1, and the phases are φ1 = 0, φ2 = π / 2, φ3 = π, and φ4 = 3π / 2.

The RF parameters determined by the RF shimming unit 213 include the amplitudes (A1, A2, A3, A4) of waveform voltages supplied to the feeding points 117a, 117b, 117c, 117d of the channels 114a, 114b, 114c, 114d, respectively. At least one of the phases (φ1, φ2, φ3, φ4).

[simulation result]
First, no measurement noise was performed in an ideal state, a conventional index (U _SD, U _NEMA) the B ₁ distribution after RF shimming with, reference to FIG. 7 (A) ~ FIG 7 (F) I will explain. Here, it is assumed that there is no noise.

FIG. 7A shows a B ₁ distribution 510 in the numerical human body model 401 without RF shimming, that is, when the initially set QD irradiation is performed as it is. Note that the B ₁ distribution 510 shown in FIG. 7A is referred to as an initial B ₁ distribution.

7 (B) shows the case of performing the RF shimming using standard deviation index U _SD of (U _SD minimized), a B ₁ distribution 520 in the voxel model 401. FIG. 7C shows the B ₁ distribution 530 in the numerical human body model 401 when RF shimming is performed using the maximum / minimum index U _NEMA (U _NEMA minimization).

7D to FIG. 7F, in order to show the degree of fat suppression, the B ₁ distribution shown in 510 to 530 is divided into a B ₁ value within a range where fat can be suppressed and a B ₁ value outside the range. It distinguishes and shows. Here, the B ₁ value region 501 in the range where fat can be suppressed is shown in white, and the B ₁ value region 502 outside the range is shown in black. The hatched area 503 is an area where the human body model 401 does not exist. In these drawings, a black region 502 is a region where the fat signal cannot be removed and the fat signal remains in the final image because the B ₁ value is outside the range in which fat can be suppressed. Hereinafter, in this specification, B ₁ distribution, and distinction and B ₁ value and the range of B ₁ value of fat suppression extent possible, the binarized figure, referred to as a binary map.

FIG. 7D is a binarized map 511 in the numerical human body model 401 without RF shimming shown in FIG. Further, FIG. 7 (E) in the case of performing RF shimming using standard deviation index U _SD, a binary map 521 in the voxel model 401. FIG. 7F is a binarized map 531 in the numerical human body model 401 when RF shimming is performed using the maximum / minimum index U _NEMA .

If without RF shimming compared with (binary map 511), who in the case of performing RF shimming (binary map 521) using a standard deviation index U _SD to index a small area of the black region 502 , B ₁ non-uniformity reduction effect is observed. However, in the binarized map 521, a black region 502 remains in a region (lumbar vertebra region) 403 where the lumbar spine exists, as in a region 504 indicated by a white arrow in the figure.

On the other hand, in the case of performing the RF shimming using a maximum minimum index U _NEMA the index (binary map 531), when performing RF shimming (binary map 521) using a standard deviation index U _SD to index In comparison, the area of the black region 502 is small. Further, the area of the black region 502 of the lumbar region 403 is reduced.

FIG. 8A to FIG. 8C show the appearance frequency (histogram) for each B ₁ value in the numerical human body model 401 in each case of FIG. 7A to FIG. 7C. FIG. 8 (A) is a histogram 610 of the case of no RF shimming, FIG. 8 (B), the histogram 620 in the case of performing RF shimming using a standard deviation index U _SD, FIG. 8 (C) Max Min It is a histogram 630 when RF shimming using the index U _NEMA is performed. In each figure, the horizontal axis indicates the B ₁ value, and the vertical axis indicates the number of points (frequency) at which the B ₁ value exists. Hereinafter, the same applies to the diagrams showing the histograms in this specification.

These histograms 610-630, compared to the case without RF shimming (histogram 610), who when performing RF shimming using a standard deviation index U _SD of (histogram 620) are in the range of B ₁ value takes It is narrower. That is, it can be said that B ₁ nonuniformity is reduced.

Further, when RF shimming is performed using the maximum / minimum index U _NEMA (histogram 630), the B ₁ value is further increased compared to when RF shimming is performed using the standard deviation index U _SD (histogram 620). The range to take is narrow.

For example, B ₁ value in between two dashed lines 601 and 602 in the figure, when it can be fat suppression, FIG. 8 (B), the RF shimming using a standard deviation index U _SD, some fat suppression There are points with B ₁ values that are not possible (B ₁ values of arrows 603 and 604 in the figure). On the other hand, in RF shimming using the maximum / minimum index U _NEMA in FIG. 8C, since all points are within the dotted line, fat suppression is possible at all points.

From the above results, in an ideal state without measurement noise, RF shimming using the maximum / minimum index U _NEMA is more effective for B ₁ maximum and B ₁ minimum than RF shimming using the standard deviation index U _SD. It can be seen that the difference between the values can be minimized and a high fat suppression effect may be expected.

However, in actual imaging, measurement noise is mixed. Here, the result when the assumed measurement noise is added to the simulation result will be described. That was done by measuring noisy state, a conventional index (U _SD, U _NEMA) the B ₁ distribution after RF shimming using, will be described with reference to FIG. 9 (A) ~ FIG 9 (F).

In this simulation, as a measurement noise, a B ₁ value which is smaller by one point than the initial B ₁ distribution 510 obtained by the simulation of initial irradiation (QD irradiation) is inserted as noise.

An initial B ₁ distribution 512 is shown in FIG. In this figure, a black point 551 is a B ₁ value inserted assuming measurement noise. The B ₁ value inserted as the measurement noise 551 is a small B ₁ value in accordance with the actual measurement noise.

Note that the region in which the measurement noise 551 is inserted is a region where the B ₁ value is originally small due to B ₁ non-uniformity, and the measurement noise is likely to be included even in actual B ₁ measurement. Therefore, it can be said that the measurement noise 551 reproduces the measurement noise that may occur in actual measurement.

FIG. 9 (B) if there is a measurement noise, i.e. of the initial B ₁ distribution 512 shown in FIG. 9 (A), in the case of performing RF shimming using standard deviation index U _SD, voxel models 401 B ₁ distribution 522. FIG. 9 (C) the B ₁ distribution to, in the case of performing RF shimming using a maximum minimum index U _NEMA, a B ₁ distribution 532 in the voxel model 401.

FIG. 9D is a binarized map 513 in the numerical human body model 401 without RF shimming, that is, with the initially set QD irradiation. Figure 9 (E) is, in the case of performing RF shimming using standard deviation index U _SD, a binary map 523 in the voxel model 401. FIG. 9F is a binarized map 533 in the numerical human body model 401 when RF shimming is performed using the maximum / minimum index U _NEMA .

Shown in FIG. 9 (D) and FIG. 9 (E), when there is no RF shimming, and binarized map 513, 523 in the case of performing RF shimming using a deviation index U _SD, as shown in FIG. 7 (D ) And the binarized maps 511 and 521 when there is no measurement noise 551 shown in FIG.

On the other hand, the binarization map 533 in the case of performing RF shimming using the maximum / minimum index U _NEMA shown in FIG. 9F is compared with the binarization map 531 of the same RF shimming shown in FIG. Thus, the black region 502 is enlarged. This indicates that the B ₁ non-uniformity is not properly reduced.

This is because the minimum value of B ₁ is used for calculating the maximum / minimum index U _NEMA . That is, when there is a measurement noise 551, extremely small B ₁ value (B ₁ value of the measured noise 551) will be included one point. In the calculation of the maximum and minimum indices U _NEMA, because this B ₁ value is employed as _a minimum value B. As described above, when an index based on the difference between the B ₁ maximum value and the B ₁ minimum value is used, the calculation of this index is greatly affected by noise. Therefore, an appropriate RF parameter cannot be calculated.

FIGS. 10A to 10C show histograms of the B ₁ value in the numerical human body model 401 in each case of FIGS. 9A to 9C. FIG. 10 (A) Histogram 611 of the case of no RF shimming, FIG. 10 (B), the histogram 621 in the case of performing RF shimming using a standard deviation index U _SD, FIG. 10 (C) Max Min It is a histogram 631 when RF shimming using the index U _NEMA is performed.

As shown in FIG. 10C, the histogram 631 when RF shimming using the maximum / minimum index U _NEMA is performed is B ₁ compared to the histogram 630 of FIG. 8C showing the result without the measurement noise 551. The range of values is widened. For this reason, the fat signal cannot be removed in a wide area and remains.

As described above, in order to effectively suppress fat, it is necessary to keep the B ₁ value within a predetermined range. Then, when the measured noise is not, RF shimming with maximum and minimum indicators U _NEMA can expect high fat suppression effect. However, this maximum / minimum index U _NEMA is easily affected by slight measurement noise, and it can be seen that a high fat suppression effect cannot always be expected when there is measurement noise.

[Simulation result by optimization index of this embodiment]
Next, the B ₁ distribution when RF shimming using the optimization index U _WSD of this embodiment is performed will be described with reference to FIGS. 11 (A) to 11 (F) and FIG.

11 (A) is in the absence of noise, the same B ₁ distribution as in the case of performing RF shimming, B ₁ distribution 520 in the voxel model 401 (FIG. 7 (B) using a standard deviation index U _SD ). FIG. 11B shows a B ₁ distribution 532 (same as B in FIG. 9C) in the numerical human body model 401 when RF shimming is performed using the maximum and minimum deviations U _NEMA in the presence of measurement noise 551. ₁ distribution). FIG. 11C shows the B ₁ distribution 542 in the numerical human body model 401 when RF shimming is performed using the optimization index U _WSD of this embodiment in the presence of measurement noise 551. Note that the measurement noise 551 was inserted in the same region as in the case of FIGS. 9 (A) to 9 (C).

FIG. 11D is a binarized map 521 (the same binarized map as FIG. 7E) in the numerical human body model 401 of the B ₁ distribution 520 in FIG. 11A. FIG. 11E is a binarization map 533 (the same binarization map as FIG. 9F) in the numerical human body model 401 of the B ₁ distribution 532 in FIG. 11B. FIG. 11F is a binarized map 543 in the numerical human body model 401 of the B ₁ distribution 542 of FIG.

As shown in FIG. 11 (F), according to RF shimming using the optimization index U _WSD of this embodiment, even in the case where there is measurement noise 551, other cases are shown in FIGS. It can be seen that the black region 502 is smaller than (E). That is, it has been shown that by using the optimization index U _WSD of the present embodiment, it is possible to reduce the region outside the fat suppression possible range.

12A to 12C show histograms of the B ₁ value in the numerical human body model 401 in each case of FIGS. 11A to 11C. FIG. 12A is a histogram 620 in the case of FIG. FIG. 12B is a histogram 631 in the case of FIG. FIG. 12C is a histogram 641 in the case of FIG. Incidentally, B ₁ value between dashed lines 601 dashed 602, a fat suppression possible B ₁ value.

As shown in these drawings, according to the optimization index U _WSD of the present embodiment, the distribution range of the B ₁ value becomes the narrowest even when the measurement noise 551 is present. That is, it can be seen that the difference between the B ₁ maximum value and the B ₁ minimum value can be minimized.

From the above results, the optimization index U of the present embodiment is weighted according to the B ₁ value and created using the value in the predetermined range on the minimum value side and the value in the predetermined range on the maximum value side. It was shown that by using _WSD , the difference between the B ₁ maximum value and the B ₁ minimum value can be minimized, and the fat suppression effect is maximized.

As described above, the MRI apparatus 100 according to this embodiment includes the transmission coil 114 having a plurality of channels that transmit a high-frequency magnetic field to the subject 103 and the nonuniformity of the B ₁ distribution that is the high-frequency magnetic field distribution in the region of interest. The high-frequency shimming section (RF shimming section) 213 for performing high-frequency shimming for determining a high-frequency parameter that is a parameter of a high-frequency magnetic field transmitted from each channel so as to correct the image An imaging unit 220, wherein the high frequency shimming unit 213 determines the high frequency parameter using an optimization index specified by the B ₁ value, and the optimization index includes the position of the imaging region and the B ₁ value It is created by weighting the B ₁ value according to at least one of the sizes.

The weighting may be performed by a weighting function. The optimization index is created to suppress fat signals. Further, the high frequency shimming unit 213 may further include an index creating unit 232 that creates the optimization index, and the index creating unit 232 may create the index by determining a weighting function for performing the weighting. .

In the conventional maximum / minimum index U _NEMA , only two points of B ₁ maximum value and B ₁ minimum value are used. For this reason, if measurement noise of values outside these two points occurs, the influence on the index is large.

However, the optimization index U _WSD of the present embodiment is calculated using a B ₁ value equal to or lower than the first threshold and a B ₁ value equal to or higher than the second threshold by weighting. Therefore, the number of points used when calculating the index is greater than when calculating the conventional maximum and minimum index _UNEMA . For this reason, the index U _WSD of this embodiment is not easily affected even when there is measurement noise.

Thus, according to the present embodiment, the RF parameter is determined so that the B ₁ uniformity is optimum. At this time, even if measurement noise is mixed, the difference between the B ₁ maximum value and the B ₁ minimum value can be minimized. For this reason, the fat suppression effect is also acquired.

Therefore, according to the present embodiment, in an MRI apparatus using a transmission coil having a plurality of channels, RF shimming that can also provide a fat suppression effect can be realized even when measurement noise is included.

<Modification of weighting function>
In the above embodiment, as shown in FIG. 4B, a weight function (mask) 310 formed so that the maximum group and the minimum value group are extracted from the B ₁ value of the imaging region is used. . That is, in the present embodiment, only the B ₁ value smaller than the first threshold 321 and the B ₁ value larger than the second threshold 322 are extracted as the weighting function w (B ₁ ) corresponding to the distribution of B ₁ values. A weighting function w (B ₁ ) 310 is used. However, the weighting function is not limited to this. Hereinafter, variations of the weighting function W (B ₁ ) will be described with reference to the drawings.

FIGS. 13A to 13E show examples of the weighting function w (B ₁ ) that can be used in this embodiment.

For example, when the magnitude of the measurement noise can be grasped, as shown in FIG. 13A, the B ₁ value is larger than the third threshold 323 and smaller than the first threshold 321 and the second A weighting function 311 that extracts B ₁ values in a range larger than the threshold and smaller than the fourth threshold 324 may be used.

The third threshold value 323 is smaller than the first threshold value 321 and the fourth threshold value 324 is larger than the second threshold value 322, and each of them is stored in the storage device 111 in advance. For these threshold values, for example, values outside the range of B ₁ values that can be taken when there is no measurement noise are selected.

By using such a weighting function 311, for example, B ₁ value and which revealed small value greatly affected by measurement noise, the clearly larger value as since B ₁ value can be excluded. Therefore, the influence of measurement noise can be reduced more than the weighting function 310 of the above embodiment.

Further, the weighting function 310 shown in FIG. 4B and the weighting function 311 shown in FIG. 13A show discontinuous changes in values. However, as shown in FIG. 13B, a function 312 whose value continuously changes may be set as a weighting function.

Further, if the relationship of the measurement accuracy and B ₁ value of B ₁ is apparent, it may determine the weighting function based on it. For example, when the measurement accuracy is worse as the B ₁ value is smaller, a weighting function that uses a smaller weight as the smaller B ₁ value is used. Examples of such weighting functions 313 and 314 are shown in FIGS. 13C and 13D.

In the weighting function 313 shown in FIG. 13C, the weight value continuously increases as the B ₁ value increases in a predetermined range, and is constant in a region larger than the range. In a small area, the shape is zero. A weighting function 314 shown in FIG. 13D has a shape that employs a B ₁ value that is equal to or greater than a predetermined threshold.

By using such a weighting function, it is possible to perform RF shimming using the B ₁ value with high measurement accuracy as a priority.

Furthermore, when RF shimming is performed using B ₁ values in a specific range, a weighting function 315 as shown in FIG. 13E may be used. For example, when it is known that the local SAR is increased due to a particularly large B ₁ value, the local SAR can be reduced by weighting using the weighting function 315 shown in FIG.

<Modification of weighting function>
In the present embodiment, the weighting function is determined according to the B ₁ value. However, it is not limited to this. For example, the weight multiplied by the B ₁ value may be a weight corresponding to the position in the ROI. In this case, the weighting function is a function having the position in the ROI as a variable. An example of the optimization index U _WSD calculation formula in this case is shown in Formula (5).

Moreover, the weight to be multiplied by the B ₁ value can be a weight according to the position and B ₁ value in the ROI. In this case, the weighting function may be a function w (B ₁ , r) having the position (spatial coordinate r) in the ROI and the B ₁ value as variables. An example of the optimization index U _WSD calculation formula in this case is shown in Formula (6).

For example, when it is desired to effectively suppress fat, information on a region where fat exists is acquired in advance. Then, the weight of the spatial coordinates of the area where fat exists is set to 1, and the weight of the spatial coordinates of the area where fat does not exist is set to 0. Thereby, RF shimming can be performed using the B ₁ value of only the region where fat exists.

Further, when it is desired to actively reduce the local SAR of a predetermined part in the subject 103, the region of the part is set to ROI. Then, the overlap value of a part of the region in the set ROI is set to 1, and in other regions, the weight is set so as to approach 0 as the distance from the set part of the region increases. As a result, the B ₁ distribution in a part of the set regions is made uniform, and the other regions can be gradually set to low signals.

<Modification of index calculation formula>
Further, the optimization index U _WSD of the present embodiment is not limited to the above calculation formula. For example, you may calculate with the following formula | equation (7).

In this case, optimization metrics U _WSD is a weighting function w (B ₁₎ 310 the B ₁ value (standard deviation of those by multiplying the B ₁ value for each position (B ₁ (r)), the weighting function w (B ₁ ) 310 but obtained by multiplying the B ₁ value (position each of B ₁ value (B ₁ (r)) obtained by dividing an average value.

Further, the optimization index U _WNEMA used in the present embodiment may be calculated according to the following equation (8). The conventional maximum / minimum index U _NEMA uses a specific B ₁ value as the maximum and minimum values, but the optimization index U _WNEMA calculated by the following equation (8) Instead of the minimum value, an average value of B ₁ values in a predetermined range is used.

Incidentally, Nupper, Nlower are each a second threshold 322 is greater than B ₁ value, the first threshold value 321 is smaller than B ₁ value in FIG. 4 (B).

That is, the optimization metrics U _WNEMA is the difference between the first threshold value 321 is smaller than B ₁ value and the second threshold 322 is greater than B ₁ value, is smaller than B ₁ value and the second first threshold 321 Divided by the sum of B ₁ values greater than the threshold 322.

<Modification of RF determination process>
Note that the optimization index U _WSD of the above embodiment may be used in combination with other indices.

Other indexes include, for example, an index related to whole body SAR, an index related to local SAR, an index related to RF irradiation power, and an index related to the average value of a high-frequency magnetic field.

When used in combination, the RF parameter determination unit 233 calculates an optimal solution using either the optimization index U _WSD of this embodiment or the combination index as an objective function and the other as a constraint.

The RF parameter determination unit 233 determines the RF parameter so as to minimize the optimization index U _WSD under a constraint condition determined using a predetermined second optimization index. At this time, the constraint conditions are, for example, setting the whole body specific absorption rate to a predetermined value or less, setting the local specific absorption rate to a predetermined value or less, minimizing the irradiation power of the high-frequency magnetic field, For example, the ratio of the average value of the high-frequency magnetic field in the region to the predetermined average value of the high-frequency magnetic field in the second region is minimized.

For example, when the index U _WSD of the present embodiment is used as the objective function and the whole body specific absorption rate is used as the constraint condition, the uniformity of the B ₁ value can be reduced while keeping the whole body SAR below a certain value.

For example, when the index U _WSD of the present embodiment is used as the objective function and the irradiation power of the high frequency magnetic field is used as the constraint condition, the uniformity of the B ₁ value is reduced while keeping the irradiation power of the high frequency magnetic field below a certain value. Can do.

For example, when the index U _WSD of the present embodiment is used as the objective function and the local specific absorption rate is used as the constraint condition, the uniformity of the B ₁ value can be reduced while keeping the local SAR below a certain value.

For example, when the index related to the whole body SAR is used as the objective function and the index U _WSD of this embodiment is used as the constraint condition, the whole body SAR can be reduced while ensuring the uniformity of the B ₁ value.

Further, when an index relating to the local SAR is used as the objective function and the index U _WSD of the present embodiment is used as the constraint condition, the local SAR can be reduced while ensuring the uniformity of the B ₁ value.

In addition, when an index related to RF irradiation power is used as the objective function and the index U _WSD of the present embodiment is used as the constraint condition, the RF irradiation power can be reduced while ensuring the uniformity of the B ₁ value.

Further, as an index used for the objective function and the constraint condition, an index combining a plurality of types of indices may be used. For example, α × U _WSD + β × whole body SAR is used as an objective function, and when local SAR is used as a constraint, the unevenness of B ₁ value is considered in consideration of the balance of whole body SAR while keeping the local SAR below a certain value. Can be reduced. Α and β are proportional coefficients.

Depending on the purpose, the types of indicators to be used (or combinations of indicator types; hereinafter simply referred to as indicator types) may be registered in the storage device 111 in advance. The RF shimming unit 213 extracts an index type used for RF parameter calculation from the storage device 111 according to a user instruction, and calculates an RF parameter.

As described above, according to the present modification, RF shimming that achieves the maximum effect can be realized in an MRI apparatus using a transmission coil having a plurality of channels even when measurement noise is included. .

It should be noted that the type of index to be registered is not the purpose but may be held in association with the imaging region. Further, the type of the index may be held in association with the imaging condition. In this case, when the imaging condition is set instead of the user giving an instruction each time, the index type used for the RF shimming is automatically extracted from the storage device 111, and the RF parameter is set accordingly. You may comprise.

<Modification of optimization method>
In the above-described embodiment, the RF parameter determination unit 233 calculates the RF parameter using a general optimization problem solving method. However, the RF parameter calculation method is not limited to this.

For example, a solution that minimizes the objective function by exhaustively changing the amplitude and phase values may be obtained. For example, the value of the objective function is calculated by changing the values of the amplitude and the phase by 1 dB and 5 degrees, respectively, and the amplitude and the phase in the case of the minimum are obtained. However, if it takes a lot of calculation time to change the amplitude and phase comprehensively, for example, find the amplitude and phase that takes the minimum value of the objective function with the amplitude and phase change amount increased at the beginning, Next, the amplitude and phase may be obtained in the vicinity of the amplitude and phase values with the amount of change reduced. The initial values of the amplitude and phase when performing these solutions are stored in the storage device 111 in advance. In addition, when the optimum amplitude and phase can be predicted to some extent in advance, the predicted value may be used as an initial value, and the amplitude and phase may be comprehensively changed only for the nearby values.

Further, the RF parameter determination unit 233 may determine only the high frequency magnetic field condition by changing only one of the amplitude and the phase.

Also, RF shimming unit 213, every time the high-frequency magnetic field conditions are changed, perform B ₁ distribution measurement for measuring the B ₁ distribution in the imaging area may be obtained B ₁ value in the imaging region.

In addition, when the RF parameter is changed comprehensively, the index creation unit 232 uses the obtained B ₁ value every time the RF parameter is changed, or the histogram shown in FIG. In addition, a B1 distribution and a binarized map may be presented to the user, and designation of the first threshold value 321 and the second threshold value 322 may be received from the user via a histogram.

In the above embodiment, the case where the imaging region is mainly two-dimensional is illustrated, but RF shimming can be performed by the same method even in the case of three-dimensional.

In the above embodiment, the 3T MRI apparatus 100 and the 4-channel transmission coil 114 have been described as examples. However, the configuration of the MRI apparatus 100 is not limited thereto. A transmission coil 114 having a higher magnetic field than 3T and a number of channels larger than 4 channels may be used.

In the above-described embodiment, the computer 109 included in the MRI apparatus 100 includes the RF shimming unit 213 and is configured to calculate at least one of the optimum RF amplitude and phase, but is not limited thereto. For example, the RF shimming unit 213 may be constructed on a computer that can transmit and receive data to and from the MRI apparatus 100 and is independent of the MRI apparatus 100.

Similarly, data necessary for each process and data generated by each process are stored not on the storage device 111 included in the MRI apparatus 100 but on an independent storage device accessible by the MRI apparatus 100 or the computer 109. May be.

Further, the method of the present embodiment can be applied to various imaging fields including medical use.

100: MRI apparatus, 101: magnet, 102: gradient coil, 103: subject, 104: sequencer, 105: gradient magnetic field power source, 106: high-frequency magnetic field generator, 107: table, 108: receiver, 109: calculator, 110: display device, 111: storage device, 112: shim coil, 113: shim power source, 114: transmission coil, 114a: channel, 114a: channel, 114b: channel, 114c: channel, 114d: channel, 115: reception coil, 117a : Feeding point, 117b: feeding point, 117c: feeding point, 117d: feeding point, 210: imaging condition setting unit, 211: imaging position setting unit, 212: static magnetic field shimming unit, 213: RF shimming unit, 220: imaging unit , 231: B ₁ distribution measuring unit, 232: index creation unit, 233: RF Pas Meter determination unit, 300: histogram of B ₁ value, 310: weighting function, 311: weighting function, 312: weighting function, 313: weighting function, 314: weighting function, 315: weighting function, 321: first threshold value, 322 : Second threshold, 323: third threshold, 324: fourth threshold, 331: range of B ₁ value, 332: range of B ₁ value, 401: numerical human body model, 402: ROI, 403: lumbar region , 501: white areas, 502: black area, 503: hatched region, 504: region, 510: B ₁ distribution 511: binarization map, 512: B ₁ distribution 513: binarization map, 520: B ₁ distribution, 521: binarization map, 522: B ₁ distribution 523: binarization map, 530: B ₁ distribution 531: binarization map, 532: B ₁ distribution 533: binarization map 542 : B ₁ distribution, 543: Binary map, 551: Measurement noise, 601: Broken line, 602: Broken line, 603: Arrow, 610: Histogram, 611: Histogram, 620: Histogram, 621: Histogram, 630: Histogram, 631 : Histogram, 641: Histogram

Claims (10)

1. A transmission coil having a plurality of channels each transmitting a high-frequency magnetic field to the subject;
   A high-frequency shimming unit that performs high-frequency shimming to determine a high-frequency parameter that is a parameter of a high-frequency magnetic field transmitted from each channel so as to correct nonuniformity of the B ₁ distribution that is a high-frequency magnetic field distribution in the region of interest;
   An imaging unit that images using a high-frequency parameter determined by the high-frequency shimming unit,
   The high-frequency shimming unit determines the high-frequency parameter using an optimization index specified by the B ₁ value,
   The optimization indicator, a magnetic resonance imaging apparatus, characterized in that in which the position and weighting according to at least one of the magnitude of the B ₁ value in the imaging region obtained by performing the B ₁ value.
2. The magnetic resonance imaging apparatus according to claim 1,
   The weighting is a magnetic resonance imaging apparatus characterized in that it is made by multiplying the weighting function to the B ₁ value.
3. The magnetic resonance imaging apparatus according to claim 1,
   The magnetic resonance imaging apparatus, wherein the optimization index is created so as to suppress fat signals.
4. The magnetic resonance imaging apparatus according to claim 2,
   The optimization index is weighted according to the B ₁ value,
   The weighting function has a shape for extracting the B ₁ value smaller than a predetermined first threshold and the B ₁ value larger than the first threshold and larger than a predetermined second threshold. A magnetic resonance imaging apparatus.
5. The magnetic resonance imaging apparatus according to claim 2,
   The magnetic resonance imaging apparatus, wherein the optimization index is obtained by dividing a standard deviation obtained by multiplying the B ₁ value for each position by the weighting function by an average value of the B ₁ values.
6. The magnetic resonance imaging apparatus according to claim 1,
   The high frequency shimming unit further includes an index creating unit that creates the optimization index,
   The magnetic resonance imaging apparatus, wherein the index creating unit creates the index by determining a weighting function for performing the weighting.
7. The magnetic resonance imaging apparatus according to claim 6,
   The optimization index is weighted according to the B ₁ value,
   The weighting function has a first threshold value smaller than the B ₁ value, the greater the first threshold value, a shape for extracting a second threshold value greater than said B ₁ value determined in advance,
   The index creation unit presents to the user a histogram of B ₁ values when irradiation is performed under a predetermined imaging condition, and accepts designation of the first threshold value and the second threshold value via the histogram. A magnetic resonance imaging apparatus.
8. The magnetic resonance imaging apparatus according to claim 4,
   The optimization metrics, the a first threshold value smaller than B ₁ value, the difference between the second threshold value is greater than B ₁ value, the a first threshold value smaller than B ₁ value, the second threshold value A magnetic resonance imaging apparatus characterized by being divided by the sum of a larger B ₁ value.
9. The magnetic resonance imaging apparatus according to claim 1,
   The high-frequency shimming unit determines the high-frequency parameter so as to minimize the optimization index under a constraint condition determined using a predetermined second optimization index,
   The constraint condition is that the whole body specific absorption rate is set to a predetermined value or less, the local specific absorption rate is set to a predetermined value or less, the irradiation power of the high frequency magnetic field is minimized, and the high frequency in the predetermined first region is set. A magnetic resonance imaging apparatus comprising: minimizing a ratio between an average value of a magnetic field and an average value of a high-frequency magnetic field in a predetermined second region.
10. Determining parameters of the high frequency magnetic field transmitted from each of the plurality of channels of the transmission coil so as to correct non-uniformity of the high frequency magnetic field distribution in the region of interest using an optimization index specified by a predetermined B ₁ value;
    The high frequency magnetic field parameters are determined to minimize the optimization index;
    The optimization index, high-frequency magnetic field shimming method characterized by being created by performing position and B ₁ value of the size weighted according to at least one of the imaging region in the B ₁ value.

Images