JP6113012B2 - Magnetic resonance imaging apparatus and method for calculating B1 map for correction - Google Patents

Description

  The present invention relates to a technique for obtaining a transmission sensitivity distribution of an RF coil that transmits an RF pulse in a magnetic resonance imaging apparatus (hereinafter referred to as an MRI apparatus).

  In recent years, due to advantages such as SNR, ultrahigh magnetic field MRI apparatuses using a main magnetic field (B0) of 3T or more are widely used. Since the Larmor frequency of the nucleus that generates the nuclear magnetic resonance signal (MR signal) is proportional to the magnetic field intensity, the RF pulse frequency increases at a magnetic field intensity exceeding 3T, and the wavelength is shortened to the same size as the size of the human body. . For this reason, a standing wave etc. generate | occur | produce and the amplitude (B1) of RF wave becomes uneven in a human body. The spatial distribution of the amplitude of the RF wave irradiated by the MRI apparatus is also called a transmission sensitivity distribution, and its uniformity becomes a problem.

  When the transmission sensitivity distribution is not uniform, the flip angle (FA) is different within the irradiation range, so that a portion having a different SNR occurs in one image. In order to eliminate the signal unevenness in the image, a technique has been developed in which a transmission sensitivity distribution is obtained from the image, RF shimming parameters are set so that the transmission sensitivity distribution is uniform, and the image is corrected.

  On the other hand, in the MR image, it is necessary to correct the influence of the sensitivity distribution of the receiving coil on the image, and a method for simultaneously correcting the reception sensitivity distribution and the transmission sensitivity distribution has been proposed (Patent Document 1, Non-Patent Document 1). In this technique, in addition to the image measured by the whole-body coil with a uniform reception sensitivity distribution and the image measured by the reception coil used for actual image measurement, the whole-body coil or the A reception sensitivity distribution and an RF wave amplitude distribution (also referred to as a B1 map) are created using an image measured using a reception coil, and signal unevenness due to the reception sensitivity distribution and the transmission sensitivity distribution is corrected. Non-Patent Document 1 describes that, when correcting signal unevenness using a B1 map, a signal intensity formula that takes into account RF shimming parameters when measuring an image to be corrected is described.

JP 2007-289344 A

C. C. Gugle, et al, “Collection of Dielectric Resonance Effect at 3T and above using B1-mapping” SMRM 13 (2005), P846.

  In normal MRI, images of a plurality of cross sections of a subject are obtained. However, when the transmission sensitivity distribution is measured under the same RF application condition for each cross section, the measurement time as a whole is extended. In particular, as described in Patent Document 1 or the like, in the technique in which correction by the reception sensitivity distribution and the transmission sensitivity distribution is performed at the same time, more data types are required before an actual image is measured, and the measurement time is further extended. To do. In addition, when calculating the correction coefficient from the acquired B1 map using the signal intensity equation, the T1 value and the T2 value of the tissue are necessary, but in order to obtain the T1 value and the T2 value for each tissue in the image. In addition, several types of measurements need to be performed, which is accompanied by further extension of measurement time.

  Further, in the case where the transmission coil is a multi-channel coil composed of a plurality of small coils, there is a problem that the accuracy is lowered when the B1 map of each small coil is combined into the B1 map of the transmission coil.

  Therefore, an object of the present invention is to provide a B1 map that can cope with correction of images of a large number of cross sections with a small measurement time. It is another object of the present invention to provide an MRI apparatus that enables highly accurate correction of signal unevenness.

  In order to solve the above problems, the present invention acquires B1 maps for a number of slices smaller than the number of slices for measurement (hereinafter referred to as main measurement) for acquiring an image to be corrected, and three-dimensionally fits the B1 map. A three-dimensional B1 map is assumed. Further, a B1 map is acquired for each of the real part and the imaginary part, and this complex B1 map is three-dimensionally fitted.

  Specifically, the MRI apparatus of the present invention includes a transmission RF coil that irradiates an RF object to an inspection object placed in a static magnetic field, a reception RF coil that receives an echo signal from the inspection object, and the reception RF. An image creation unit that creates an image related to the inspection object using an echo signal received by the coil, and the image creation unit includes a transmission sensitivity distribution calculation unit that calculates a transmission sensitivity distribution of the transmission RF coil, The transmission sensitivity distribution calculation unit includes a two-dimensional B1 map calculation unit that calculates a two-dimensional complex transmission sensitivity distribution using echo signals respectively obtained from a plurality of cross sections to be inspected, and an actual two-dimensional complex transmission sensitivity distribution. A three-dimensional B1 map calculation unit that calculates a three-dimensional complex transmission sensitivity distribution of a volume including the plurality of cross sections for each of the part and the imaginary part, and the three-dimensional B1 map calculation unit And calculating a correction B1 map from the three-dimensional complex transmission sensitivity distribution of the real part and the imaginary part and a correction B1 map calculation unit.

  In the MRI apparatus of the present invention, the transmission RF coil is a multi-channel coil composed of a plurality of small coils, and the transmission sensitivity distribution calculation unit calculates the three-dimensional transmission calculated for each of the plurality of small coils. A synthesis unit for synthesizing the sensitivity distribution using a predetermined RF shim parameter is provided.

  The method for calculating the B1 map for correction according to the present invention is a method for calculating the B1 map for correction using echo signals respectively obtained from a plurality of cross sections to be inspected, and is a two-dimensional complex using the echo signal. A transmission sensitivity distribution is calculated, a three-dimensional complex transmission sensitivity distribution of the volume including the plurality of cross sections is calculated for each of a real part and an imaginary part of the two-dimensional complex transmission sensitivity distribution, and a three-dimensional complex of the real part is calculated. The correction B1 map is calculated from the transmission sensitivity distribution and the three-dimensional complex transmission sensitivity distribution of the imaginary part.

  Since the measurement for obtaining the B1 map is a measurement with a smaller number of slices than the actual measurement, it is possible to prevent the measurement time from being extended. Since the B1 map is a complex B1 map and each is subjected to three-dimensional fitting, and the B1 map (absolute value) is obtained, the fitting accuracy is high, and a highly accurate B1 map can be obtained.

  Even if the measurement conditions (number of slices, slice position, etc.) of the main measurement change, it is only necessary to cut out a desired position of the three-dimensional B1 map, and remeasurement is unnecessary. Furthermore, even when the RF shimming parameters (amplitude and phase) are changed in this measurement, the correction coefficient can be calculated using the changed RF shimming parameters, so that it is not necessary to perform B1 measurement or fitting again. .

Overall configuration diagram of an MRI apparatus to which the present invention is applied Functional block diagram of the image creation unit The flowchart which shows the process sequence of embodiment of the MRI apparatus of this invention The flowchart which shows the detail of step S233 of FIG. (A) is a figure explaining extrapolation area reservation step S301 of FIG. 4, (b) is a figure explaining fitting step S305, S306 of FIG. FIG. 5A is a diagram showing a fitting result, FIG. 5A is a diagram showing a B1 map (absolute value) of each channel measured for the RF transmission coil, and FIG. 5B is a diagram showing a B1 map (absolute value) of each channel after fitting; (C) is a figure which shows the phase of each channel after fitting. The flowchart which shows the detail of step S235 of FIG. The flowchart which shows the detail of step S237 of FIG. The figure which shows the relationship between effective FA and the signal strength for every structure | tissue, (a) And (b) is a graph which shows the relationship between the effective FA and signal strength in 2D sequence and 3D sequence, (c) is T1 value and T2 of a structure | tissue It is a table | surface which shows a value and the change width of the signal strength in a typical effective FA range. The figure explaining correction coefficient calculation of a second embodiment Flow chart showing the processing procedure of the third embodiment It is a figure explaining the outline | summary of the process of 3rd embodiment, (a) is a graph which shows the relationship between a fitting and a mask area | region, (b) is a figure which shows the update of the fitting result by a measured value, (c) is a necessity for filtering It is a figure explaining the property, (d) It is a figure explaining the concept of distance.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing an overall outline of an MRI apparatus to which the present invention is applied.
The MRI apparatus shown in FIG. 1 includes a static magnetic field generation unit 2 having a static magnetic field generation magnet (not shown), a gradient magnetic field generation unit 3 that gives a magnetic field gradient to the static magnetic field generated by the static magnetic field generation magnet, and a static magnetic field generation unit 2. An RF transmitter 5 for transmitting an RF pulse to the subject 1 placed in a static magnetic field space in which a magnetic field is generated, an RF receiver 6 for receiving a magnetic resonance signal (echo signal) generated from the subject 1, and a gradient magnetic field A sequencer 4 that controls the operation of the generator 3, the RF transmitter 5, and the RF receiver 6, and a signal that performs predetermined signal processing on the magnetic resonance signal received by the RF receiver 6 and forms an image of the subject 1 A processing unit 7, a control unit 8 that controls each unit described above, and an input unit 9 for an operator to input commands and data necessary for the operation of the apparatus are provided.

  The static magnetic field magnet provided in the static magnetic field generating unit 2 may be a permanent magnet, a normal conducting magnet, or a superconducting magnet, and is an open type or cylindrical shape in which the magnet is arranged above and below the space (measurement space) where the subject is placed. Any of a cylindrical type in which a measurement space is provided in the magnet bore may be used. The subject 1 is usually carried into the measurement space while being laid on a bed.

  The gradient magnetic field generation unit 3 includes three sets of gradient magnetic field coils (Gx, Gy, Gz) 31 that form magnetic field gradients in three axial directions orthogonal to each other, and is connected to the gradient magnetic field power supply 32 and set in the sequencer 4. The operation is performed according to the operation timing (imaging sequence).

  The RF transmitter 5 includes a high-frequency transmitter 11 that generates a high frequency of a predetermined frequency, a modulator 12 that modulates a high-frequency signal generated by the high-frequency transmitter 11 according to a command from the sequencer 4, a high-frequency amplifier 13, and an RF transmission coil 14a. Become. The RF receiving unit 6 includes a quadrature phase detector 16 and an AD converter 17 that quadrature-detects the signals received by the RF receiver coil 14b, the amplifier 15, and the RF receiver coil 14b into two systems of signals.

  The RF transmission coil 14a and the RF reception coil 14b are arranged in the vicinity of the subject 1 in the measurement space, and one high-frequency coil may serve as the transmission coil and the reception coil, or one or more separate high-frequency coils are provided. May be. In addition, one or both of the RF transmitting coil and the RF receiving coil may be a multi-channel coil including a plurality of small coils. In this case, the RF transmission unit 5 can be configured to include a switching unit (not shown) that switches multi-channels, and the RF reception unit 6 has the same number of RF reception units 6 as the number of channels of the RF reception coil. It can be set as the structure provided with. In the MRI apparatus of the present embodiment described below, the RF transmission coil can also serve as a reception coil, and is a multichannel coil, and includes a plurality of RF reception units 6.

  The signal processing unit 7 performs signal processing such as correction calculation and image reconstruction and image processing on the digitized reception signal. The CPU also functions as the control unit 8 and memories such as the ROM 21 and the RAM 22. And an external memory such as a magnetic disk 18 and an optical disk 19 and a display 20 for displaying an image, a GUI or the like as a processing result. The CPU also serves as the control unit 8. The processing performed by the signal processing unit 7 further includes calculation of a transmission sensitivity distribution of the RF transmission coil 14a that is a multi-channel coil, correction of a reconstructed image using the transmission sensitivity distribution, and the like.

  The control unit 8 controls the operation of each unit via the sequencer 4 according to the command input from the input unit 9. In the MRI apparatus of this embodiment, measurement (main measurement) for obtaining an image of a subject and measurement for obtaining correction information for correcting an image obtained by the main measurement are performed separately. The measurement for obtaining the correction information is a measurement for obtaining a transmission sensitivity distribution of the RF transmission coil, and is referred to as B1 map measurement. The pulse sequences for performing the main measurement and the B1 map measurement are pre-programmed and stored in the memory in the control unit 8, and parameters necessary for actually operating these pulse sequences are input to the input unit 9 It is input from.

  A functional block diagram of the image creating unit 70 among the functions of the signal processing unit 7 is shown in FIG. As shown in the figure, the image creation unit 70 includes an image reconstruction unit 71, a reception sensitivity distribution calculation unit 72, a transmission sensitivity distribution calculation unit 73, an image correction unit 74, and functions, coefficients, and calculations necessary for the calculations of the units 71 to 74. A memory unit 75 for storing results and the like is provided. The memory unit 75 can be realized by a memory such as a ROM, a RAM, an optical disk, or a magnetic disk shown in FIG.

  The transmission sensitivity distribution calculation unit 73 includes a two-dimensional B1 map calculation unit 731, a fitting unit 732, a three-dimensional B1 map calculation unit 733, a correction coefficient calculation unit 734, and the like.

  Based on the outline of the MRI apparatus described above, an operation of the MRI apparatus of the present embodiment, mainly an embodiment of processing in the image creating unit will be described.

<First embodiment>
The operation flow of the MRI apparatus of this embodiment is shown in FIG. As shown in FIG. 3, the operation is roughly divided into a step S210 for obtaining an image (before correction) 201 of the subject, a reception sensitivity correction step S220, and a transmission sensitivity correction step S230. Step S210 includes step S211 for performing the main measurement and step S212 for reconstructing an image using the received RF signal measured by the main measurement. Receiving sensitivity correction step S220, the step S221, step S222 for obtaining the sensitivity image using the received RF signal measured in step S221 of performing measurements to obtain the sensitivity image image, step S223 of calculating the reception sensitivity correction coefficient 202 from the sensitivity image Step S224 for adjusting the size of the reception sensitivity correction coefficient 202 to the pre-correction image and Step S225 for performing reception sensitivity correction by applying the reception sensitivity correction coefficient 202 are included.

The transmission sensitivity correction step S230 includes a step S231 for performing measurement for obtaining the two-dimensional transmission sensitivity distribution 203 for each channel of the RF transmission coil, a step S232 for calculating a complex two-dimensional B1 map using the measured signals, and 2 Step S233 for calculating a three-dimensional B1 map by fitting from the three-dimensional B1 map, and a transmission sensitivity distribution at the same slice position as the pre-correction image 201 is cut out from the three-dimensional B1 map 205, and a B1 map for each channel is synthesized and a B1 map for correction is synthesized. and step S234 of creating a flop, and the step S235 of calculating a correction B1 map or we transmit sensitivity correction coefficient 203, and step S236 to match the size of the transmission sensitivity correction coefficient 203 in the correction image before further transmission sensitivity correction coefficient 203 Step S237 for performing transmission sensitivity correction using No.

In FIG. 3, the last step S225 of the reception sensitivity correction step S220 and the last S237 of the transmission sensitivity correction step S230 are collectively described as an image correction step.
Either step S220 or step S230 may be performed first, and the main measurement step S210 may be performed before the last correction step S224 of step S220 and step S233 of step S230.

Hereinafter, specific processing contents of each step will be described.
<Main measurement step S210>
This measurement is performed by executing an imaging sequence according to the purpose of measurement under predetermined imaging conditions (imaging parameters, imaging cross section, number of slices, etc.) (S211). The RF shimming parameters (RF amplitude and phase) in the imaging sequence of the main measurement are stored in the memory for use in the subsequent calculation in step S230. The echo signal obtained by the execution of the imaging sequence is reconstructed by an operation such as Fourier transform and stored in the memory as an uncorrected image (S212).

<Reception sensitivity correction step S220>
The reception sensitivity correction is performed by using a known method described in, for example, Magnetic Resonance of Medicine 53 (2005), p666. Measuring the sensitivity image image using the coil for the same RF receiver and RF receiving coil used in this measurement is first (S221). The sensitivity image is measured with a lower spatial resolution than the main measurement on the same cross section as the main measurement.

  Next, a low-pass filter or the like is applied to the measured sensitivity image, the structure of the tissue is removed, and only the reception sensitivity distribution is extracted to obtain a reception sensitivity map (S222). The reception sensitivity correction coefficient 202 can be created by applying a process of converting the correction coefficient into a correction coefficient such as taking the reciprocal of the reception sensitivity map (S223).

The correction coefficient 202 is interpolated and enlarged to the same size as the pre-correction image 201 (S224). This process is a process to match the size of the sensitivity image image matrix size is different from the pre-correction image 201, to increase the size by interpolating between neighboring pixels of the sensitivity image image image values of the surrounding. Finally, multiplied by the correction coefficient after size adjustment obtained in step S224 to the before-correction image 201 (S225). By this processing, signal unevenness due to the reception sensitivity distribution is corrected. When the reception sensitivity correction S220 is performed after the transmission sensitivity correction S230, the correction coefficient 202 obtained in step S223 is stored in the memory, and when the transmission sensitivity distribution correction is performed in the subsequent step S230, it is read out from the memory and received. Make corrections together.

<Transmission sensitivity correction step S230>
<< Steps S231 and S232 >>
First, a pulse sequence for B1 map measurement is executed for each channel of the RF transmission coil (S231). A publicly known method can be used for B1 map measurement. For example, a double angle method (DAM) for measuring B1 by calculating an image captured using an RF pulse having an arbitrary flip angle and an image captured using an RF pulse having a double flip angle, pre-pulse application A method of calculating the B1 distribution by taking the ratio of the image acquired immediately after and the image acquired without applying a pre-pulse, image data using a set of pulse sequences with different TRs with the same flip angle RF pulse And a method of calculating a two-dimensional B1 distribution (Actual Flip Angle method: AFI) using the ratio of the signal of the image data and the ratio of TR can be employed.

  The B1 map measurement may require a smaller number of slices than the pre-correction image, and the measurement position may not match the pre-correction image.

  The transmission sensitivity distribution image obtained by the measurement in step S231 represents the magnetic field strength (vector) in each pixel constituting the image. In this embodiment, the magnetic field intensity, which is a complex number, is divided into a real part and an imaginary part, and a two-dimensional B1 map (complex number B1 map) is calculated for each (S232).

<< 3-dimensional B1 map calculation S233 >>
Next, by applying 3D fitting by a polynomial to the 2D B1 map of each transmission channel, the data between the slices is filled and the B1 map is extrapolated to the outside of the subject to obtain the 3D B1 map. create. Details of the three-dimensional B1 map calculation S233 will be described with reference to FIG.

[Ensure extrapolation area S301]
In this process, in order to extrapolate the complex B1 map of each transmission channel in three dimensions, first, a memory area of X × Y × Z × 2 (real part / imaginary part) × [number of transmission channels] is secured as an extrapolation area. (S301). At this time, the sizes of X, Y, and Z are determined from the FOV of the complex B1 map, the number of matrices, and the extrapolated volume. For example, when the B1 map is set to FOV: 500 mm, 64 × 64 matrix on the XY plane, and the extrapolated volume is set to 500 × 500 × 500 mm ³ , the extrapolated area is 64 if the Z direction is set to 10 mm per matrix. It becomes x64x50x2 matrix. One matrix in the Z direction preferably has a resolution comparable to that of the matrix in the X / Y direction from the viewpoint of maintaining correction accuracy. The extrapolated volume must cover the effective field of view of the device.

[Data arrangement S302]
Next, the complex B1 map of each transmission channel is arranged at a correct position in the extrapolation area (S302). For example, as shown in FIG. 5A, the extrapolation area in the Z direction is 50 matrix, the extrapolation length is −250 to 250 mm, and the Z coordinate of the complex B1 map is Z = −250, −200, −150. ,..., 200 [mm], the complex B1 map is arranged at memory positions k = 0, 5, 10,. The memory position of the extrapolation area where the complex B1 map is not arranged is extrapolated by the next fitting process. FIG. 5 shows only one data of a real part or an imaginary part of one channel.

[Phase correction S303]
Next, prior to the fitting process, phase correction of the complex B1 map is performed (S303). By performing the B1 map phase correction process before the fitting process, the fitting accuracy can be further improved.

  In order to correct the phase of each transmission channel complex B1 map, a reference phase distribution is first created. The reference phase distribution is obtained from a reference distribution obtained by QD combining each transmission channel complex B1 map. Specifically, as shown in Expression (1) and Expression (2), an offset phase (for example, 0, −90, −180, 90 [deg.]) Corresponding to the channel arrangement from the B1 map of each transmission channel. After subtracting each, the data of all channels are synthesized.

式中、ｒは画像内の位置ベクトル、ｎは送信チャンネル番号（１≦ｎ≦Ｎ、Ｎは送信チャンネル数）、ＯｒｇＲｅａｌ_ｎとＯｒｇＩｍａｇ_ｎとはｎチャンネル目の実部及び虚部のＢ１マップデータ、Ｒｅａｌ_ｎとＯｒｇＩｍａｇ_ｎはｎチャンネル目のオフセット位相減算後の実部と虚部のＢ１マップデータ、θｎはｎチャンネル目のオフセット位相、ＳｔａｎｄａｒｄＲｅａｌとＳｔａｎｄａｒｄＩｍａｇとは全チャンネル合成後の基準分布（実部と虚部）をそれぞれ示す（以下、同じ）。なお、オフセット位相θｎは、ＲＦシムパラメータを用いても良い。

Where r is a position vector in the image, n is a transmission channel number (1 ≦ n ≦ N, N is the number of transmission channels), OrgReal _n and OrgImag _n are the B1 map data of the real part and imaginary part of the n-th channel , Real _n and OrgImag _n are the B1 map data of the real part and the imaginary part after the offset phase subtraction of the n-th channel, θn is the offset phase of the n-th channel, and StandardReal and StandardImag are the reference distribution (real part) And the imaginary part) (hereinafter the same). The offset phase θn may use an RF shim parameter.

Next, a phase is obtained from this reference distribution by the equation (3), and this is set as a reference phase StandardPhase.

式中、ＲｅａｌＣｏｒｒｅｃｔｅｄ_ｎとＩｍａｇｅＣｏｒｒｅｃｔｅｄ_ｎとは位相補正後の各送信チャンネル複素Ｂ１マップ（実部と虚部）をそれぞれ示す。 The reference phase obtained as described above is subtracted from the complex B1 map of all channels as shown in Equation (4).

In the formula, it RealCorrected _n and ImageCorrected _n and each transmission channel complex B1 map after the phase correction is the (real and imaginary parts), respectively.

[Mask processing S304]
Next, a mask for fitting is created (S304). The mask for fitting distinguishes a subject area and a region other than the subject so that a region other than the subject with much noise is not used for fitting. Specifically, an area where the signal value is equal to or greater than a predetermined threshold is set as an in-mask area, a fitting process is performed using the in-mask area, and values in other areas are extrapolated. For example, the threshold is set to, for example, 10% of the maximum value from the absolute value of the reference distribution described above, and the mask is created. The threshold value may be determined by another method such as a discriminant analysis method.
[Three-dimensional polynomial fitting S305, S306]

  Finally, for each transmission channel, three-dimensional polynomial fitting is performed using only real part data and only imaginary part data in the mask (S305 and S306). By fitting the real part and the imaginary part of each transmission channel B1 map three-dimensionally, the fitting accuracy can be increased and the final correction coefficient can be made highly accurate. In the three-dimensional fitting, each pixel value in an area (slice) where there is no actual measurement data is used to set an optimal fitting function using information on pixels adjacent or close to the X direction, Y direction, and Z direction of the pixel. For example, a quadratic function as shown in Expression (5) can be used.

式中、ａ〜ｊは各項の係数を示す。
係数は最小二乗法などの公知の手法で最適化する。

In the formula, a to j indicate coefficients of the respective terms.
The coefficient is optimized by a known method such as a least square method.

  Note that the fitting order in each of the three-dimensional directions may be set to be smaller in the Z direction than in the X direction and the Y direction by using a characteristic that the B1 map is not distributed greatly in the Z direction (B0 direction). . The orders in the X direction and the Y direction may also be made independent of each other.

  By performing the three-dimensional polynomial fitting as described above, three-dimensional B1 map data is obtained as shown in FIG. Since this three-dimensional B1 map has a data arrangement reflecting the actual B1 map interval, the B1 map can be obtained with high accuracy even for slice positions from which the B1 map has not been acquired.

[Update with actual data S307]
Of the obtained three-dimensional B1 map data, for the area in the mask (area whose value is higher than the predetermined threshold), the memory is overwritten so that the actual value is used instead of the fitting result. Update (S307).

[S308]
The above-described processing of S304 to S307 is performed on the B1 map of the real part and imaginary part of all channels (S308). FIG. 6 shows a B1 map that is a result of performing the above-described fitting on the 4-channel RF transmission coil. 6A shows the B1 map (absolute value) 601 of each channel measured in step S231, and FIGS. 6B and 6C show the fit in the mask after fitting the B1 map 601 for each channel. These are the absolute value 602 and phase value 603 of the B1 map obtained by updating the region with actual values. From the B1 map of FIG. 6A, it can be seen that there is a change in distribution due to the arrangement of transmission channels. When this was fitted for each channel, as shown in (b) and (c), the distribution was correctly reproduced, and the phase was also correctly reproduced because the fitting of the real part and the imaginary part was performed.

<< Correction B1 Map Creation S234 >>
Returning to FIG. 3, the B1 map creation step S234 will be described. First, a two-dimensional B1 map corresponding to the measurement position of the pre-correction image 201 is cut out from the three-dimensional B1 map obtained in the fitting step, and the complex of each transmission channel that matches the FOV and the number of slices of the pre-correction image 201 is extracted. Create a B1 map. Next, the complex B1 map of each transmission channel is complex-synthesized with an RF shimming parameter to create a correction B1 map. Details of the synthesis process (S234) will be described with reference to FIG.

[RF shim parameter setting S701]
The B1 map measured in step S231 is based on the RF shim parameter at the time of B1 map measurement. In order to correct the image 201 of the main measurement, the B1 map measured by the RF shim parameter at the time of the main measurement is used. Need to be converted to Therefore, first, the parameters used during the B1 map measurement (S231) and the amplitude ratio _An and phase difference Φ _n of the parameters used in the main measurement (S211) are defined as shown in Equation (6) (S701).

式中、Ａｍｐ_０［ｎ］、φ_０［ｎ］は、それぞれ、Ｂ１マップ計測時に使用したｎチャンネル目のＲＦシムパラメータ（絶対値、位相）、Ａｍｐ［ｎ］、φ［ｎ］は、本計測で使用したｎチャンネル目のＲＦシムパラメータ（絶対値、位相）をそれぞれ示す。

In the equation, Amp ₀ [n] and φ ₀ [n] are the RF shim parameters (absolute value and phase) of the n-th channel used at the time of B1 map measurement, and Amp [n] and φ [n] The RF shim parameters (absolute value, phase) of the nth channel used in the measurement are shown.

[Synthesis S702 to S704]
Next, each channel B1 map (extracted from the three-dimensional B1 map) is multiplied by the amplitude ratio _An and phase difference Φ _n defined by the equation (6), added for all channels, and each channel B1 map is synthesized. (S702 to S704).

式中、ＦＯｒｇＲｅａｌ_ｎとＦＯｒｇＩｍａｇ_ｎとはｎチャンネル目のＢ１マップデータ（実部と虚部、３次元Ｂ１マップから切り出したもの）、ＲｅａｌＳｙ_ｎとＩｍａｇＳｙ_ｎとは合成後の補正用Ｂ１マップデータ（実部と虚部）をそれぞれ示す。

Wherein, B1 map data of n-th channel and FOrgReal _n and FOrgImag _n (real part and imaginary part, which cut out from the three-dimensional B1 maps), RealSy _n and corrective B1 map data after combining the ImagSy _n ( Real part and imaginary part) are shown respectively.

式中、ＡｂｓＳｙｎは、補正用Ｂ１マップ（絶対値）を示す。 [Absolute value S705]
Finally, the B1 map data of the real part and the imaginary part are converted into absolute values according to the equation (8) to obtain a correction B1 map (S705).

In the formula, AbsSyn indicates a correction B1 map (absolute value).

<< Transmission Sensitivity Distribution Correction S235-S236 >>
The main measurement image 201 is corrected using the correction B1 map (absolute value) after the channel synthesis (FIG. 3, S235 to S237). Details of this step will be described in detail with reference to FIG.

[Effective FA distribution calculation S801]
First, using formula (9), an effective FA distribution is created from the correction B1 map and the set FA value (S801). The correction B1 map is “AbsSyn (r)” calculated by the equation (8) in step S234. The set FA value is a flip angle of the RF transmission coil 14a given to the RF transmission unit 5 by the sequencer 4 (FIG. 1) in this measurement.

式中、ＦＡは実効ＦＡ分布、Ｂ１Ｍａｐは補正用Ｂ１マップ（＝式（８）のＡｂｓＳｙｎ（ｒ））、ＦＡ_０は設定ＦＡ値をそれぞれ示す。

In the equation, FA represents an effective FA distribution, B1Map represents a correction B1 map (= AbsSyn (r) in equation (8)), and FA ₀ represents a set FA value.

式中、ＦＡは実効ＦＡ、ＴＲは設定ＴＲ値、ＴＥは設定ＴＥ値、Ｔ１は縦緩和時間、Ｔ２は横緩和時間である（以下、同じ）。 [Signal intensity distribution calculation]
Next, the signal intensity distribution is calculated using the effective FA distribution calculated by Expression (9) and the signal intensity expression determined by the imaging pulse sequence of the main measurement (S802). As the signal intensity equation, for example, equation (10) can be used for a GrE (gradient echo) sequence, and equation (11) can be used for an SE (spin echo) sequence.

In the formula, FA is an effective FA, TR is a set TR value, TE is a set TE value, T1 is a longitudinal relaxation time, and T2 is a lateral relaxation time (hereinafter the same).

式中、ＦＡは実効ＦＡ（励起パルスのフリップ角）分布、ＲＦＡは再収束パルスの実効フリップ角分布をそれぞれ示す。

In the equation, FA indicates an effective FA (excitation pulse flip angle) distribution, and RFA indicates an effective flip angle distribution of refocusing pulses.

  When using the signal intensity formula described above, it is necessary to obtain the T1 value and the T2 value for each cross section. In the present embodiment, the effective FA and the similarity of the signal intensity change of each tissue are used to represent the cross section. T1 and T2 of typical tissues are used. The similarity between the effective FA and the signal intensity change of each tissue will be described with reference to FIG. FIGS. 9A and 9B show the signals of liver, spleen, pancreas, paraspinal muscle, and subcutaneous fat in 2D-GrE and 3D-GrE sequences using typical TR and TE, respectively. It is a figure which shows the relationship between intensity | strength and effective FA.

  In the figure, a range surrounded by vertical lines on both sides of the three vertical lines (range indicated by arrows) indicates a range of set FA ± 20%, and is a range that is predicted to become an effective FA in actual measurement (hereinafter referred to as “the FA”). , Referred to as effective FA range). From the graphs (a) and (b), the change width of the signal intensity of each tissue in this effective FA range (the ratio of the change to the intensity at the set FA) is read together with the T1 value and the T2 value (c) It is shown in the table. As shown in the table, although T1 and T2 differ depending on the tissue, it can be seen that the change width of the signal intensity in the effective FA range is almost the same. For this reason, it can be seen that even if the change in signal intensity (signal unevenness) is estimated from the T1 value and T2 value of typical tissues and applied to all tissues, there is no significant error. Therefore, in the present embodiment, representative T1 values and T2 values are used as the T1 value and the T2 value of the signal intensity expressions such as Expression (10) and Expression (11). Typical tissues include white matter for the head, liver and muscles for the abdomen, and muscles for the joints. As a result, the calculation accuracy of the signal strength equation is ensured and it is not necessary to measure the T1 value and the T2 value of each tissue, so that the measurement time can be shortened.

式中、ＳＩは信号強度分布、ＦＡは実効ＦＡ分布、Ｓｉｇｎａｌは信号強度式をそれぞれ示す。 The calculation of the signal intensity distribution (S802) can be performed from the expression (12) using the signal intensity expression described above.

In the equations, SI represents a signal intensity distribution, FA represents an effective FA distribution, and Signal represents a signal intensity expression.

式中、Ｃｏｅｆは送信補正係数を示す。 [Correction coefficient calculation S803 (S235)]
Finally, using equation (13), the inverse of the signal intensity distribution SI is taken to obtain a transmission sensitivity correction coefficient (S803).

In the equation, Coef represents a transmission correction coefficient.

<< S236, S237 >>
The correction coefficient Coef is interpolated and enlarged to the same size as the pre-correction image 201 (S236), and multiplied by the image to which the reception sensitivity correction is applied (S237). The increase in the size of the correction coefficient is the same as the increase in the size of the correction coefficient for the reception sensitivity correction.

Through the above processing, a sensitivity-corrected image 205 obtained by performing signal unevenness correction by the reception sensitivity distribution and signal unevenness correction by the transmission sensitivity distribution is obtained. The sensitivity correction image 205 is obtained by multiplying the pre-correction image 201 by the reception sensitivity correction coefficient 202 and the transmission sensitivity correction coefficient 203 at the slice position, as shown in Expression (14), and receives the reception sensitivity distribution correction S 220 and the transmission. It can be seen that the sensitivity-corrected images obtained by performing any processing of the sensitivity distribution correction S 230 are the same.

式中、Corrected Iは感度補正画像、OrgIは補正前画像、ＲCoefは受信感度補正係数、TCoeは送信感度補正係数を示す。

In the equation, Corrected I indicates a sensitivity correction image, OrgI indicates a pre-correction image, RCoef indicates a reception sensitivity correction coefficient, and TCoe indicates a transmission sensitivity correction coefficient.

  According to the MRI apparatus of the present embodiment, the B1 map measurement can be performed with a smaller number of slices than the image of the main measurement, so the time for the B1 map measurement can be shortened. In addition, since the three-dimensional B1 map is obtained, it is not necessary to remeasure the B1 map even if the number of slices and the slice position for the main measurement change. Further, by dividing the B1 map of each channel into a real part and an imaginary part and performing a three-dimensional polynomial fitting, a highly accurate B1 map can be obtained. In particular, the accuracy of fitting can be improved by performing phase correction prior to fitting.

  Furthermore, according to the MRI apparatus of this embodiment, even when the RF shimming parameter is changed in the main measurement, the B1 map may be synthesized using the changed RF shimming parameter, and the fitting process needs to be performed again. There is no.

<Second embodiment>
This embodiment is characterized in that a condition based on the effective FA is added to the correction coefficient in the transmission sensitivity distribution correction as a function of the signal processing unit based on the MRI apparatus of the first embodiment. Other configurations and processing procedures are the same as those in the first embodiment, and different points will be mainly described below.

  Also in the present embodiment, the configuration of the signal processing unit (image creating unit 70) is the same as the functional block diagram shown in FIG. 2, and the processing procedure of the image creating unit is the same as the flow shown in FIG. However, step S235 shown in FIG. 3 (step S803 in FIG. 8) is different. Hereinafter, the processing of this embodiment will be described in detail with reference to FIGS. 3 and 8.

  In step S235, first, an effective FA distribution is calculated from the correction B1 map and the set FA value in order to correct the main measurement image 201 using the correction B1 map (absolute value) after channel synthesis (FIG. 8). : S801). Next, the signal intensity distribution is calculated using the effective FA distribution and the signal intensity formula determined by the imaging pulse sequence of this measurement (S802). Finally, the reciprocal of the signal intensity distribution is obtained by Expression (13) to obtain a transmission sensitivity correction coefficient (S803).

  In the reciprocalization step S803, if the effective FA distribution approaches 0, the obtained result is likely to be incorrect. There are two cases where the effective FA is extremely small. One is when the correction B1 map is very small, and the other is when the set FA itself is very small. Of these, when the setting FA is extremely small, the SNR of the obtained image is low, so that it may be considered that it is not actually used. On the other hand, when the B1 map for correction is extremely small, this can actually occur when the B1 non-uniformity is extremely large or when an error is included in the B1 map measurement.

  In general, in the GrE-based sequence, the signal strength approaches 0 when the effective FA is extremely small from the signal strength equation of Equation (10). A graph showing the relationship between the effective FA and the signal intensity is shown in FIG. In the figure, a graph 1001 is a signal intensity calculated from the signal intensity equation. In the reciprocalization step S803, when the transmission sensitivity correction coefficient is calculated using this signal strength, the correction coefficient that is the reciprocal of the signal strength distribution diverges, resulting in an incorrect result.

  Therefore, in the present embodiment, the correction factor is calculated by replacing the effective FA with the maximum signal strength as the boundary value and replacing the signal strength with the signal strength at the boundary value at a position where the effective FA is equal to or less than the boundary value. A graph 1002 in FIG. 10 shows the signal strength replaced with the signal strength of the boundary value at a position where the effective FA is equal to or lower than the boundary value.

The boundary FA is an FA known as the Ernst angle, and is calculated by the equation (15), so that it is uniquely obtained from the TR and T1 values.

  In the MRI apparatus of this embodiment, when calculating the correction coefficient in the transmission sensitivity distribution correction, the polarity of the signal strength with respect to the FA can be lowered to the right by treating the effective FA below the Ernst angle as the Ernst angle. Even when an error is included in the effective FA, a large error is not caused in the correction coefficient.

<Third embodiment>
The present embodiment is characterized in that a filtering function for the three-dimensional B1 map after fitting is added as a function of the image creation unit based on the MRI apparatus of the first embodiment. Other configurations are the same as those of the first embodiment, and different points will be mainly described below.

  FIG. 11 shows a processing flow of the image creating unit according to the present embodiment. The same processes as those in FIG. 4 are denoted by the same reference numerals. As can be seen from the comparison with the processing flow of the first embodiment shown in FIG. 4, in this embodiment as well, the two-dimensional B1 map of each channel after phase correction is fitted after masking and the fitting result is updated in the mask. (S301 to S307) The same as in the first embodiment, but in this embodiment, a discontinuous step that may occur between the in-mask region and the other region (extrapolated region) after updating in the mask. The process (S901, S902) which eliminates is performed.

  In particular, when the B1 map has a high-order distribution in the chest or abdomen, a boundary may be formed between the area updated in the mask and the extrapolated area in the above-described processing S307, resulting in a discontinuous distribution. is there. This is shown in FIG. FIG. 12A shows a result 1211 obtained by fitting an area 1201 or higher area (in-mask area) 1201 and extrapolating the other area (extrapolated area) 1202, and FIG. 12B shows the fitting result 1211. Of these, the position where the measurement data of the in-mask area 1201 exists indicates the result 1212 of replacement with the measurement data. A portion surrounded by a dotted line in FIG. 12B is a part of the fitting result 1212 including a boundary between the in-mask region 1201 and the extrapolated region 1202 indicated by a one-dot chain line, and only that portion is shown in FIG. Yes. (C) shows a discontinuity between the in-mask area 1201 and the extrapolated area 1202.

In the present embodiment, in order to eliminate this discontinuity and realize highly accurate correction, filtering according to the distance from the in-mask region is applied. First, the distance from the mask area is set to the pixels of the entire B1 map (S901). Specifically, as shown in FIG. 12 (d), set the distance 1 to the pixels in the mask area, then set the distance 2 to pixels adjacent to the distance 1 pixel. By repeating this operation while incrementing the distance to 3, 4,..., The distance is set for all pixels in the extrapolation region.

  Next, a filtering process is performed using a smoothing filter having different radii depending on the set distance (S902). As the smoothing filter, a Gaussian filter or the like is used in order to suppress discontinuity at the boundary inside and outside the mask. The filter may be created in three dimensions, and the filter radius in the X / Y / Z direction may be set independently. At this time, the filter radius corresponding to the distance is used by using the distance from the in-mask region calculated in step S901. For example, the filter radius is increased at a position where the distance from the mask area is small and affected by discontinuity, and conversely, the filter radius is decreased at a position where the distance from the mask area is large and hardly affected by discontinuity. The filter radius control according to the distance is shown in Equation (16).

式中、Filter Radiusはフィルタ半径、Ｄは距離、Ｄ_０は距離上限、ａとｂはフィルタ半径の最大値と最小値をそれぞれ示す。このような３次元フィルタを用いて不連続を抑制することができ、Ｂ１マップを実測していない位置においても、精度の高いＢ１マップを作成できる。

In the equation, Filter Radius is a filter radius, D is a distance, D ₀ is a distance upper limit, and a and b are maximum and minimum values of the filter radius. Discontinuity can be suppressed using such a three-dimensional filter, and a B1 map with high accuracy can be created even at a position where the B1 map is not actually measured.

  As described above, each embodiment of the present invention has been described. However, in the MRI apparatus of the present invention, the features of each embodiment can be combined as appropriate, and non-essential processing can be omitted as appropriate. For example, in the present invention, the transmission sensitivity distribution correction of the main measurement image is preferably performed together with the reception sensitivity distribution correction, but the case where only the transmission sensitivity distribution correction is performed is also included in the present invention. In the above-described embodiment, the case where the RF transmission coil is a multi-channel coil including a plurality of small RF reception coils has been described. However, the present invention can also be applied to an RF transmission coil including a single coil.

  Furthermore, in the above-described embodiment, the case where the correction coefficient for correcting the image of the main measurement is obtained using the three-dimensional B1 map calculated by the transmission sensitivity distribution calculation unit has been described. However, the present invention is applied to an optimum technique of RF shimming parameters. It is also possible.

  According to the present invention, highly accurate transmission sensitivity distribution correction can be performed without causing an increase in measurement time. In particular, even if there is a change in imaging conditions or RF shimming parameters in this measurement, the measurement for correcting the transmission sensitivity distribution can be made unnecessary again, greatly increasing the burden on both the subject and the operator. Can be reduced.

2 ... static magnetic field generator, 3 ... gradient magnetic field generator, 4 ... sequencer, 5 ... RF transmitter, 6 ... RF receiver, 7 ... signal processor, 8 ..CPU (control unit), 9 ... input unit, 70 ... image creation unit, 71 ... image reconstruction unit, 72 ... reception sensitivity distribution calculation unit, 73 ... transmission sensitivity distribution calculation Part, 74... Image correction part.

Claims (15)

A transmission RF coil for irradiating an RF pulse to an inspection object placed in a static magnetic field, a reception RF coil for receiving an echo signal from the inspection object, and the inspection object using an echo signal received by the reception RF coil An image creation unit for creating an image related to
The image creation unit includes a transmission sensitivity distribution calculation unit that calculates a transmission sensitivity distribution of the transmission RF coil, and the transmission sensitivity distribution calculation unit 2 uses echo signals respectively obtained from a plurality of cross sections of the inspection target. A two-dimensional B1 map calculation unit for calculating a three-dimensional complex transmission sensitivity distribution, and a three-dimensional complex transmission sensitivity distribution of a volume including the plurality of cross sections for each of a real part and an imaginary part of the two-dimensional complex transmission sensitivity distribution And a correction B1 map calculation unit for calculating a correction B1 map from the real and imaginary three-dimensional complex transmission sensitivity distributions calculated by the three-dimensional B1 map calculation unit. A magnetic resonance imaging apparatus. The magnetic resonance imaging apparatus according to claim 1,
The transmission RF coil is a multi-channel coil composed of a plurality of small coils,
The transmission sensitivity distribution calculation unit includes a synthesis unit that synthesizes a three-dimensional transmission sensitivity distribution calculated for each of the plurality of small coils using a predetermined RF shim parameter. . The magnetic resonance imaging apparatus according to claim 1 or 2,
The magnetic resonance imaging apparatus, wherein the image creation unit includes an image correction unit that corrects the created image using the correction B1 map calculated by the correction B1 map calculation unit. The magnetic resonance imaging apparatus according to claim 3,
The transmission sensitivity distribution calculation unit calculates a transmission sensitivity correction coefficient using the correction B1 map calculated by the correction B1 map calculation unit and the signal intensity formula under the RF application condition set in the measurement of the correction target image. A correction coefficient calculation unit
The magnetic resonance imaging apparatus, wherein the image correction unit corrects the image using a transmission sensitivity correction coefficient calculated by the correction coefficient calculation unit. The magnetic resonance imaging apparatus according to claim 4,
The signal intensity formula used by the correction coefficient calculation unit includes the T1 value and the T2 value of the examination target, and the correction coefficient calculation unit indicates the similarity between the signal intensity change of each tissue to be examined and the RF shim parameter. A magnetic resonance imaging apparatus characterized in that the signal intensity equation is determined by using. The magnetic resonance imaging apparatus according to claim 4,
The magnetic resonance imaging apparatus, wherein the correction coefficient calculation unit calculates a transmission sensitivity correction coefficient using the Ernst angle as the RF flip angle when the RF flip angle of the RF application condition is smaller than the Ernst angle. The magnetic resonance imaging apparatus according to claim 1,
The magnetic resonance imaging apparatus, wherein the three-dimensional B1 map calculation unit includes a fitting unit that performs three-dimensional fitting on each of a real part and an imaginary part of the two-dimensional complex transmission sensitivity distribution. The magnetic resonance imaging apparatus according to claim 7,
The three-dimensional B1 map calculation unit includes a phase correction unit that corrects the phase of the two-dimensional complex transmission sensitivity distribution for each cross section, and the fitting unit performs two-dimensional complex transmission after phase correction by the phase correction unit. A magnetic resonance imaging apparatus that performs fitting on a real part and an imaginary part of a sensitivity distribution. The magnetic resonance imaging apparatus according to claim 7 or 8,
The fitting unit excludes a region that is not used in the fitting process from the two-dimensional complex transmission sensitivity distribution, fits a region that is not excluded, and extrapolates the fitting result to the excluded region, thereby obtaining a three-dimensional complex transmission sensitivity. A magnetic resonance imaging apparatus characterized by obtaining a distribution. A magnetic resonance imaging apparatus according to any one of claims 7 to 9,
The fitting unit assigns a value of the position corresponding to the two-dimensional complex transmission sensitivity distribution to a three-dimensional complex transmission sensitivity distribution by substituting the value of the two-dimensional complex transmission sensitivity distribution in the fitting result. A magnetic resonance imaging apparatus. The magnetic resonance imaging apparatus according to claim 10,
The magnetic resonance imaging apparatus, wherein the fitting unit further includes a smoothing unit that smoothes a three-dimensional complex transmission sensitivity distribution. The magnetic resonance imaging apparatus according to claim 3,
The image creation unit further includes a reception sensitivity distribution calculation unit that calculates a sensitivity distribution of the reception coil,
The magnetic resonance imaging apparatus, wherein the image correction unit corrects the image using the three-dimensional transmission sensitivity distribution and the reception sensitivity distribution calculated by the reception sensitivity distribution calculation unit. A method of calculating a correction B1 map using echo signals respectively obtained from a plurality of cross sections to be inspected,
Using the echo signal to calculate a two-dimensional complex transmission sensitivity distribution;
Calculating a three-dimensional complex transmission sensitivity distribution of a volume including the plurality of cross sections for each of a real part and an imaginary part of the two-dimensional complex transmission sensitivity distribution;
A method for calculating a correction B1 map from a three-dimensional complex transmission sensitivity distribution of a real part and a three-dimensional complex transmission sensitivity distribution of an imaginary part. 14. A method according to claim 13, comprising:
The transmission RF coil is a multi-channel coil composed of a plurality of small coils,
For each of the plurality of small coils, the three-dimensional complex transmission sensitivity distribution is calculated, and the three-dimensional complex transmission sensitivity distribution of each small coil is synthesized using the RF shim parameter, and the synthesized three-dimensional complex transmission is obtained. A method of calculating a correction B1 map from a sensitivity distribution. 15. A method according to claim 13 or 14, comprising
Further, a transmission sensitivity correction coefficient is calculated using the B1 map for correction and a signal intensity formula under RF application conditions set in measurement of the correction target image.

Images