JP6783642B2 - Sensitivity map calculation method for magnetic resonance imaging device and receiving coil - Google Patents

Description

The present invention relates to a magnetic resonance imaging apparatus (hereinafter referred to as an MRI apparatus), and more particularly to a technique for calculating an accurate sensitivity distribution of a receiving coil of the MRI apparatus.

In the MRI apparatus, a receiving RF coil (hereinafter referred to as a receiving coil) is used to receive a nuclear magnetic resonance signal which is a high frequency magnetic field. The receiving coil includes a body coil having a wide sensitivity distribution and used for whole body imaging, and a surface coil (also referred to as a local coil) having a high sensitivity distribution in a narrow region and suitable for local imaging. A general MRI apparatus has a body coil built into the gantry as standard, and information obtained by imaging using the body coil is also used in a prescan for determining imaging conditions.

The image obtained by the MRI apparatus is affected by the sensitivity distribution of the receiving coil. In particular, the surface coil has high sensitivity but is inferior in the uniformity of the sensitivity distribution. Therefore, the image after imaging is corrected (sensitivity correction) using the sensitivity information of the receiving coil in order to eliminate the influence of the sensitivity distribution of the receiving coil.

Accurate sensitivity information is required to improve the accuracy of sensitivity correction. The sensitivity information of the receiving coil is obtained from, for example, a homogeneous image of the phantom (that is, the image itself is the sensitivity information), but the sensitivity of the receiving coil is when the subject (for example, the human body) is present and when the phantom is present. It is different from when using. In order to determine the sensitivity of the receiving coil under the condition that the subject is present, the sensitivity of the receiving coil is obtained by image processing such as reconstruction and filtering using only the data in the low frequency region using the data obtained by imaging the subject. There is a technique for obtaining a distribution (self-reference type correction) (for example, Patent Documents 1 to 3). Patent Document 3 discloses a technique for obtaining a reception sensitivity distribution by filtering a proton-weighted image corrected using a B1 map as an improved technique for self-reference type correction.

Further, since the sensitivity of the surface coil is non-uniform, it has been proposed to perform a reference scan and correct the sensitivity by the ratio with the relatively uniform body coil (Patent Document 4).
Since the ratio of the sensitivity distribution of the surface coil to the sensitivity distribution of the body coil can be obtained by a reference scan of the subject, it is possible to correct the sensitivity unevenness of the body coil to be relatively uniform when the image is taken with the surface coil. However, when the sensitivity distribution of the body coil is not uniform, it is necessary to correct the sensitivity distribution of the body coil. At that time, if the sensitivity distribution of the body coil is obtained by a method such as self-reference, it may fail depending on the contrast of the reference image.

Further, various techniques have been proposed for correcting the sensitivity non-uniformity of the body coil that remains after the correction using the body coil (Patent Documents 5 to 8). For example, Patent Document 5 proposes a method of estimating an initial value of an image using images obtained from a plurality of coils and applying a homomorphic filter and a multi-channel blind deconvolution to obtain a sensitivity distribution. ..

Further, Patent Document 6 employs a technique for correcting the sensitivity distribution information of the surface coil by using an image obtained by the body coil which can be regarded as having a substantially uniform sensitivity distribution in the same region as the surface coil.

Further, in the technique described in Patent Document 6, the non-uniformity of the body coil is corrected by image processing. Further, a technique has also been proposed in which the sensitivity distribution of the body coil to be referred to is corrected, and then the sensitivity distribution of the surface coil is obtained by using the sensitivity distribution of the corrected body coil. For example, Patent Document 7 discloses a technique for correcting the reception sensitivity of a receiving coil by making the B1 distribution of the transmitting coil uniform by shimming on the premise that the transmitting sensitivity and the receiving sensitivity are equal.
Patent Document 8 discloses a technique for obtaining T2 by imaging a plurality of TEs, estimating the structure and its proton density from T2, creating a reference image in TE at the time of imaging from T2 and the proton density, and obtaining a sensitivity map. However, this technique requires measurement of T2 and may be affected by various errors (correspondence between T2 and proton density, calculated T2 value).

Japanese Patent Application Laid-Open No. 2007-503239 (US Patent Application Publication No. 2006/0261809) U.S. Pat. No. 6,680,610 U.S. Pat. No. 9,316,707 EP069594A1 (1996/02/07) U.S. Patent Application Publication No. 2013/0281822 Japanese Patent Application Laid-Open No. 2009-207756 (US Patent Application Publication No. 2009/0224756) Patent No. 5706899 (US Pat. No. 9036884) Patent No. 413405

In the present invention, the sensitivity distribution correction of the body coil or the like is not affected by the contrast of the reference image, and the sensitivity distribution (sensitivity map) of the receiving coil in the presence of the subject can be calculated accurately, thereby providing sensitivity information. An object is to improve the image quality of the subject image created by using the image. The sensitivity distribution calculation method of the present invention can be applied not only to the body coil but also to the surface coil.

In order to solve the above problem, in the present invention, for a receiving coil having a plurality of receiving channels, the relationship between the sensitivity maps of the plurality of receiving channels acquired other than the inspection target is obtained as an index (first index). The relationship between the images (reference images) of multiple reception channels acquired for the inspection target is obtained as the second index, and the map of the second index and the map of the first index are transformed into similar distributions. Calculate the coordinate conversion formula to be used. The sensitivity map of the receiving coil acquired by other than the inspection target is corrected by the coordinate conversion formula to create a reception sensitivity map substantially equivalent to the reception sensitivity map for the inspection target. The receiving coil used for imaging, that is, the receiving coil for creating the receiving sensitivity map may be a part or all of the receiving coil having a plurality of receiving channels used in calculating the coordinate conversion formula, or may be a different coil. ..

Further, when the reception sensitivity map of the wide area receiving coil is created, the sensitivity map of a different receiving coil used for imaging the subject may be further corrected by the received sensitivity map of the created wide area receiving coil.

That is, the MRI apparatus of the present invention includes a receiving coil having a plurality of receiving channels, receives the nuclear magnetic resonance signal of the inspection target, and acquires the image of the inspection target, and the sensitivity information of the receiving coil. It is provided with a calculation unit that performs calculations necessary for creating an image of the inspection target by using the sensitivity map of the receiving coil acquired in advance other than the inspection target, a sensitivity map of each of the plurality of receiving channels, or the plurality of receptions. A storage unit that stores a map derived from the sensitivity map of each channel is further provided, and the calculation unit includes a first index map showing the relationship between the sensitivity maps of the plurality of receiving channels, and the imaging unit is the inspection target. Calculates a coordinate conversion formula for coordinate conversion of the first index map so as to minimize the difference from the second index map representing the relationship between the plurality of reference images acquired for each reception channel of the reception coil. It is characterized by including a coordinate conversion formula calculation unit and a sensitivity map creation unit that creates a reception sensitivity map when the inspection target exists by using the coordinate conversion formula.
The map derived from the sensitivity map of each of the plurality of receiving channels may include, in addition to the first index map, data in the process of calculating the first index map from the sensitivity map.

According to the present invention, by using the sensitivity map prepared in advance of the receiving coil having a plurality of receiving channels and the coordinate conversion formula of the sensitivity map calculated by using the reference image, the contrast of the reference image is not affected and the accuracy is high. A reception sensitivity map can be created.

The figure which shows the whole structure of the example of the MRI apparatus to which this invention is applied. The flow chart which shows the outline of processing of the MRI apparatus of embodiment. The functional block diagram of the signal processing part of the MRI apparatus of 1st Embodiment. The flow chart which shows an example of the processing of the signal processing part of 1st Embodiment. The figure which shows typically the outline of the coordinate transformation process in 1st Embodiment. The flow chart which shows the other example of the processing of the signal processing part of 1st Embodiment. The figure explaining the effect of processing by 1st Embodiment. The functional block diagram of the signal processing unit of the 2nd Embodiment. The flow chart which shows the processing flow of the signal processing part of the 2nd Embodiment. The figure which shows (a) the initial sensitivity map obtained by simulation, (b) the sensitivity map obtained by phantom imaging, and (c) the initial sensitivity map after correction. The functional block diagram of the signal processing unit of the third embodiment. The flow chart which shows the processing flow of the signal processing part of 3rd Embodiment. The figure which shows the combination example of the receiving coil in 4th Embodiment. The flow chart which shows the processing flow of the signal processing part of 4th Embodiment.

First, an overview of the MRI apparatus to which the present invention is applied will be described with reference to the drawings. FIG. 1 is a block diagram showing an overall configuration of an embodiment of an MRI apparatus according to the present invention. This MRI apparatus obtains a tomographic image of a subject by utilizing an NMR phenomenon, and includes a static magnetic field generating unit 2, a gradient magnetic field generating unit 3, a transmitting unit 5, a receiving unit 6, and a signal processing unit 7. , The sequencer 4 and the central processing unit (CPU) 8 are provided.

The static magnetic field generator 2 generates a uniform static magnetic field in the space around the subject 1, and includes a permanent magnet type, a normal conduction type, or a superconducting type static magnetic field generator. Depending on the direction of the generated magnetic field, there are a vertical magnetic field method, a horizontal magnetic field method, and the like, and the present invention can be applied to any of the methods.

The gradient magnetic field generating unit 3 includes a gradient magnetic field coil 9 that applies a gradient magnetic field in the three axes of X, Y, and Z, which is a coordinate system (static coordinate system) of the MRI apparatus, and a gradient magnetic field that drives each gradient magnetic field coil. It is composed of a power supply 10, and by driving the gradient magnetic field power supply 10 of each coil according to a command from the sequencer 4, the gradient magnetic fields Gx, Gy, and Gz are applied in the three axial directions of X, Y, and Z. By combining these triaxially gradient magnetic fields, a gradient magnetic field in any direction can be generated. For example, at the time of imaging, a slice direction gradient pulse (Gs) is applied in a direction orthogonal to the slice plane (imaging cross section) to set a slice plane with respect to the subject 1, and the rest orthogonal to the slice plane and orthogonal to each other. A phase-encoding direction gradient pulse (Gp) and a frequency-encoding direction gradient pulse (Gf) are applied in the two directions to encode the position information in each direction in the echo signal.

The sequencer 4 is a control means that repeatedly applies a high-frequency magnetic field pulse (hereinafter referred to as “RF pulse”) and a gradient magnetic field pulse in a predetermined pulse sequence, operates under the control of the CPU 8, and collects data of a tomographic image of the subject 1. Various commands necessary for the above are sent to the transmitting unit 5, the gradient magnetic field generating unit 3, and the receiving unit 6.

The transmission unit 5 irradiates the subject 1 with an RF pulse in order to cause nuclear magnetic resonance in the nuclear spins of the atoms constituting the biological tissue of the subject 1. The high-frequency oscillator 11, the modulator 12, and the high-frequency amplifier 13 and a high frequency coil (transmission coil) 14a on the transmission side are provided. The RF pulse output from the high-frequency oscillator 11 is amplitude-modulated by the modulator 12 at the timing commanded by the sequencer 4, and the amplitude-modulated RF pulse is amplified by the high-frequency amplifier 13 and then placed close to the subject 1. By supplying the transmission coil 14a, an RF pulse is applied to the subject 1.

The receiving unit 6 detects an echo signal (NMR signal) emitted by nuclear magnetic resonance of the nuclear spins constituting the biological tissue of the subject 1, and is a high-frequency coil (reception coil) 14b and a signal amplifier 15 on the receiving side. The orthogonal phase detector 16 and the A / D converter 17 are provided. The NMR signal of the response of the subject 1 induced by the electromagnetic wave emitted from the transmitting coil 14a is detected by the receiving coil 14b arranged close to the subject 1, amplified by the signal amplifier 15, and then amplified by the signal amplifier 15. The signals are divided into two systems orthogonal to each other by the orthogonal phase detector 16 at the timing according to the above command, each of which is converted into a digital quantity by the A / D converter 17 and sent to the signal processing unit 7.

The signal processing unit 7 performs various data processing, display and storage of processing results, and a part of the functions are realized by the CPU 8. The processing performed by the signal processing unit 7 (CPU 8) includes image reconstruction of the subject 1, calculation of numerical values indicating the characteristics of the subject, correction of signals and processing results using the characteristics of the device such as reception coil sensitivity, and the like. included.

The signal processing unit 7 includes an external storage device 18 such as an optical disk and a magnetic disk, a display 19 including a CRT and the like, and an operation unit 20 including a trackball or a mouse and a keyboard. When the data from the receiving unit 6 is input to the CPU 8, the CPU 8 executes processing such as signal processing and image reconstruction, displays the tomographic image of the subject 1 as a result on the display 19, and is an external storage device. Record on 18 magnetic disks or the like. The CPU 8 functions as a calculation unit that performs the above-described processing, and also functions as a control unit that controls the sequencer 4 and the entire device.

The operation unit 20 inputs various control information of the MRI apparatus and control information of processing performed by the signal processing unit 7. The operation unit 20 is arranged close to the display 19, and the operator interactively passes through the operation unit 20 while looking at the display 19. Controls various processes of the MRI apparatus.

In FIG. 1, the high-frequency coil 14a and the gradient magnetic field coil 9 on the transmitting side face the subject 1 in the static magnetic field space of the static magnetic field generating unit 2 into which the subject 1 is inserted, if the vertical magnetic field method is used. If it is a horizontal magnetic field method, it is installed so as to surround the subject 1. Further, the high frequency coil 14b on the receiving side is installed so as to face or surround the subject 1.

The transmitting coil 14a and the receiving coil 14b may be separate high-frequency coils, but one high-frequency coil may also serve as the transmitting coil 14a and the receiving coil 14b. In addition, a whole-body coil that also serves as transmission / reception and a high-frequency coil such as a local coil other than the coil may be used in combination. The MRI apparatus to which the present invention is applied includes a receiving coil having a plurality of receiving channels as at least one high frequency coil. "Having a plurality of receiving channels" means that the receiving coil is a combination of a plurality of small receiving coils, and each small receiving coil is a receiving coil that receives an NMR signal, for example, a multi-array coil. The purpose is to include all of having a plurality of receiving ports, and each receiving port is a receiving coil for receiving an NMR signal, for example, a bird cage type coil or the like. Hereinafter, these are collectively referred to as "multi-channel receiving coil" or "multi-channel coil".

Next, an outline of imaging and signal processing of the MRI apparatus having the above-described configuration will be described with reference to the flow of FIG. In the following description, the static magnetic field generating unit 2, the gradient magnetic field generating unit 3, the sequencer 4, the transmitting unit 5, and the receiving unit 6 are collectively referred to as an imaging unit 100.

First, the subject 1 is placed in the static magnetic field space of the MRI apparatus, and the imaging unit 100 is driven to perform imaging. The imaging includes a reference scan (S101) for positioning and the like and a main imaging (S103), and at least the reference scan uses a multi-channel coil, for example, a whole body coil. The signal processing unit 7 creates a reference image using the NMR signal acquired by the reference scan (S103), and creates a target image using the NMR signal obtained by the main imaging (S104). The target image includes a proton density image, a diffusion-weighted image and other enhanced images, and a calculated value image derived from these images. The image creation step S104 includes correction of sensitivity unevenness (hereinafter, simply referred to as sensitivity correction) (S1042) using a sensitivity map of the receiving coil, which is acquired or calculated in advance.

In the MRI apparatus of the embodiment described in detail below, the sensitivity map of the multi-channel coil and the reference scan data acquired in advance are used in the calculation of the sensitivity map of the receiving coil. Specifically, the coordinate conversion formula is derived using the sensitivity map of each channel and the reference image of each channel obtained by the reference scan, and this coordinate conversion formula is applied to receive the receiving coil when the subject is present. Create a sensitivity map (S1041). The signal processing unit 7 creates an image by performing calculations necessary for sensitivity correction (S1042) and image reconstruction using this reception sensitivity map.

The function as the calculation unit of the signal processing unit 7 described above can be realized by executing software such as a memory (storage unit) provided in the CPU 8 and a program incorporated in the CPU 8. Further, some of the functions may be realized by hardware such as ASIC and FPGA.
Based on the above outline, each embodiment of the processing performed by the signal processing unit 7 will be described below.

<First Embodiment>
In this embodiment, a coordinate conversion formula is created using the data obtained from the multi-channel coil, and a sensitivity map in the presence of the subject is created. The present embodiment further corrects the sensitivity of the surface coil used in the present imaging by using the sensitivity map created for the multi-channel coil.

As shown in the functional block diagram of FIG. 3, the signal processing unit 7 in the present embodiment includes a sensitivity calculation unit 30 that performs various calculations related to sensitivity information of the receiving coil, and an image creation unit 40 that creates an image from an NMR signal. including. The sensitivity calculation unit 30 further examines the difference in the distribution of the sensitivity map of each channel from the sensitivity map storage unit (storage unit) 31 that stores information corresponding to the sensitivity map for each channel of the multi-channel coil acquired in advance. The sensitivity map coordinate conversion determination unit 32 that determines the coordinate conversion that aligns with the difference in the distribution of the sensitivity map inherent in the reference image obtained by the reference scan for, and the sensitivity map of the coil acquired in advance are converted by the coordinate conversion. The sensitivity map creation unit 33 for creating a sensitivity map when a subject is present (referred to as a reception sensitivity map to distinguish it from a sensitivity map acquired in advance) is provided. The image creation unit 40 includes a sensitivity correction unit 41 that performs sensitivity correction using the reception sensitivity map created by the sensitivity map creation unit 33. The sensitivity map coordinate conversion determination unit 32 includes an index calculation unit 321 and a coordinate conversion formula calculation unit 322.

As shown in FIG. 4, the sensitivity maps of the multi-channel coil stored in the sensitivity map storage unit 31 are the sensitivity map Sw of the entire multi-channel coil and the sensitivity map Sn of each channel constituting the multi-channel coil (n is n). Indicates that it is the second channel, and includes an integer with the maximum number of channels). These sensitivity distributions are obtained by acquisition with a phantom or the like different from the subject to be imaged or created by computer simulation or the like, and are different from the sensitivity map when the subject is set.

Next, the details of the processing of the signal processing unit 7 will be described. FIG. 4 is a diagram showing a processing flow of the signal processing unit 7. Here, as a premise, in step S101 of FIG. 2, a reference scan using a multi-channel coil in which the sensitivity map is stored in the sensitivity map storage unit 30. Is performed, and it is assumed that a reference image is created from the NMR signals obtained in each channel by the reference scan (S101).

Using this reference image, the signal processing unit 7 (sensitivity map coordinate conversion determination unit 32) determines the coordinate conversion so that the difference in the distribution of the sensitivity map Sn of each channel is aligned with the reference scan for the subject. Specifically, first, the index calculation unit 321 calculates the relationship (first index map) between the corresponding pixels of the sensitivity map of each channel stored in the sensitivity map storage unit 31 (S401). The relationship between the corresponding pixels is an index of the relationship of the sensitivity distribution of each channel. For example, the ratio of the pixel values of each channel, the vector having the pixel value as a vector element, the declination of the vector, etc. are used. Can be done. By calculating the index for all the pixels of the sensitivity map, a map (index map) using the index as the pixel value can be obtained. In FIG. 4, step S401 for calculating the first index map is described as being performed after the reference image is created. However, since the first index map can be calculated if there is a sensitivity map for each channel, imaging is performed. The index calculation unit 321 may calculate the value prior to the above and store it in the sensitivity map storage unit 30 in advance.

The index calculation unit 321 calculates the relationship between the corresponding pixels (second index map) of the reference image of each channel by using the reference image of each channel created in S102 (S402). The index used here is the same as the index of the first index map. That is, if the first index is a vector, it is a vector, and if it is a vector declination, it is a vector declination. Subsequently, the coordinate conversion formula calculation unit 322 obtains a coordinate conversion formula that matches the first index map calculated in S401 with the second index map calculated in S402 (S403).

The coordinate conversion formula calculation method will be described by taking the case of a 2-channel coil as an example. FIG. 5 schematically shows the coordinate conversion formula calculation process. Sensitivity map storage unit _S 1 sensitivity maps of the 30 channels stored in 1 (ch1) (r), the sensitivity map of the channel 2 (ch2) and _S 2 (r). “R” indicates the coordinates of the pixel. The declination φ ₁₂ (r) of the vector S ₁₂ (here, a two-dimensional vector) having the pixel values of each of these channels as elements is calculated. The declination φ is obtained for each pixel. This is an argument distribution (first index map) that represents the relationship between the corresponding pixels of the sensitivity map.

Further, the reference image of channel 1 (ch1) is R ₁ (r), the reference image of channel 2 (ch2) is R ₂ (r), and the declination θ ₁₂ (deviation angle θ ₁₂ of the vector R ₁₂ having the pixel value of each channel as an element). r) is calculated. The declination θ is obtained for each pixel. This is an argument distribution (second index map) that represents the relationship between the pixels of the sensitivity map.

Next, the coordinate conversion formula that minimizes the difference between the declination φ and the declination θ is obtained according to the following equation. Specifically, for example, in the case of translation and scaling in the xy plane, the four parameters of translation and scaling (movement amount in the x and y directions, enlargement ratio in the x and y directions) are set to one pixel. The search is performed while changing the corresponding amount, and the position where the equation (1) is minimized is determined.
[Number 1]
Σ | (φ ₁₂ (r) −θ ₁₂ (r')) ² | (1)

式（２’）はｒをｘ−ｙ座標で表した時のアフィン変換を示し、拡大・縮小、平行移動、及び回転を含む。 The coordinate conversion formula can be expressed by, for example, formula (2) or formula (2').

Equation (2') shows the affine transformation when r is expressed in xy coordinates, and includes scaling, translation, and rotation.

The above description is an example of 2 channels, but 3 channels or more may be used, and in the case of n (n is an integer of 2 or more) channels, the angle formed by the n-dimensional vector having the sensitivity of each channel as an element becomes 0. Coordinate conversion should be performed so as to be. At that time, in order to absorb the overall deviation of the channel sensitivity, the coefficient formed by changing the coefficient applied to the sensitivity of each channel may be changed to approach 0. Further, the above-mentioned coordinate transformation is an example, and in addition, a known registration technique, rigid registration, non-rigid registration, or the like may be adopted for the coordinate transformation. For example, registration may be performed after obtaining the ratio of pixel values between channels and the declination of a vector, or registration may be performed as data having a plurality of channels as a vector. Furthermore, the target data format can be changed arbitrarily, such as by standardizing the magnitude of the vector and then performing registration.

When the coordinate conversion formula F (r → r') is determined in step S403, the sensitivity map Sw of the multi-channel coil stored in the sensitivity map storage unit 30 is coordinate-converted using this coordinate conversion formula. A sensitivity map S'w (reception sensitivity map) including the reception sensitivity of the subject is obtained (S404).

Next, the sensitivity map of the surface coil is obtained by multiplying the sensitivity map C as a ratio to the sensitivity map of the multi-channel coil obtained by deriving the coordinate conversion formula obtained in advance for the surface coil by the reception sensitivity map S'w obtained in S404. Obtain and correct the image acquired in the main imaging (S405: sensitivity correction).

The sensitivity correction (S405) of the image I obtained in the main imaging is the sensitivity as a ratio between the corrected sensitivity map S'w and the sensitivity map of the multi-channel coil from which the coordinate conversion formula of the receiving coil used in the main imaging is derived. It is performed by the following equation using the map C.
[Number 3]
I'(x, y) = I (x, y) / (C x S'w) (3)

Note that "C x S'w" in the formula (3) may be calculated before the calculation of the formula (3), or the image I'after the correction of the formula (3) may be calculated at once. Good. Further, in order to prevent the equation (3) from diverging, a process for not setting the denominator on the right side to zero, for example, a process of excluding the region where the sensitivity becomes zero from the calculation or a process of inserting a constant value is performed. May be good.

In the above example, the surface coil was used for the main imaging, and the sensitivity map C of the surface coil as the ratio to the sensitivity map of the multi-channel coil from which the coordinate conversion formula was derived was obtained by the reference scan. However, in the case of the same receiving coil as the multi-channel coil from which the coordinate conversion formula is derived, as shown in FIG. 6, the main imaging is performed in the same manner as the conventional sensitivity correction using the receiving sensitivity map S'w obtained in S404. The sensitivity of the image obtained in (S406) is corrected.

In the above example, the sum of squares of the difference in declination at each point on the sensitivity map was minimized, but instead of equation (3), offset so that the average values are aligned in the area to be compared as in equation (4). After adding, a method of minimizing the sum of squares of the difference in declination at each point on the sensitivity map may be adopted.
[Number 4]
Σ ｜ ((φ ₁₂ (r) −Avg (φ ₁₂ (r))) − (θ ₁₂ (r') −Avg (θ ₁₂ (r')))) ² ｜
(4)
Here, Avg represents the average in the area to be compared.

In the above example, the declination is used as an index showing the relationship between the sensitivity maps of each channel or between the reference images, but if the two indexes (the first index and the second index) are the same, Not limited to the declination angle, for example, another index such as a sensitivity map of each channel or a ratio or difference of pixel values of a reference image can be used.

Next, the effect of the correction according to the first embodiment will be described with reference to FIG. 7. FIG. 7 is an imaging result of the head in which the imaging conditions of the reference scan are set so that contrast is generated. The lower side is the captured image, and the upper side is a diagram showing the profile on the profile line on the image. Further, in FIG. 7, the left (a) is an image without correction, and the center (b) is an image with sensitivity corrected by a conventional method (a method in which a reference image of a body coil is subjected to a low-pass filter and used as a sensitivity map of the body coil). The right side (c) is an image whose sensitivity has been corrected by the method of the first embodiment (a method of transforming a sensitivity map acquired by a phantom in advance to obtain a sensitivity map of a body coil).

As can be seen from these results, it can be seen that the result (a) without correction has a difference in luminance value where the line profile is originally symmetrical. In the result (b) of the conventional method, the influence generated by the reference scan is not corrected, and the line profile shows that the difference between the left and right remains. On the other hand, in the result (c) of the present embodiment, the same reference scan as in the conventional method (b) is used, but it can be seen that the difference between the left and right is eliminated.

According to the present embodiment, the first index showing the relationship between the sensitivity distributions of each channel of the multi-channel coil and the second index showing the relationship between the reference images obtained in each channel of the same coil are used. A coordinate conversion formula to be matched is obtained, and a reception sensitivity map of the receiving coil including the sensitivity of the subject is created using the coordinate conversion formula. Thereby, the accuracy of the sensitivity correction can be improved.

Further, according to the present embodiment, when creating a sensitivity map of a multi-channel coil, the sensitivity map is less likely to be affected by the contrast of the reference scan. This is because it is possible to take an index that cancels the contrast, such that the contrast disappears by division in the ratio between the channels of the index used in the example. Therefore, since the imaging parameters of the reference scan can be set without worrying about the contrast, it is possible to adjust the TE of the reference scan to reduce the imaging sound.

In the above embodiment, an example of performing sensitivity correction using the sensitivity map of the receiving coil corrected by the coordinate conversion formula or the sensitivity map of another receiving coil corrected by the sensitivity map of the entire image has been described, but other It may be combined with sensitivity correction. For example, in both 2D imaging and 3D imaging, sensitivity correction is performed using the reception sensitivity map including the subject sensitivity of the present embodiment as a sensitivity map in the direction perpendicular to the body axis, and the sensitivity correction in the body axis direction is the conventional sensitivity correction. You may go there.

Further, in the above-described embodiment, the reception sensitivity correction has been described, but it goes without saying that other corrections such as transmission sensitivity correction and filter processing and image processing may be performed as in the conventional reception sensitivity correction technique.

<Second embodiment>
Also in this embodiment, a coordinate conversion formula for matching the first index calculated from the sensitivity map of each receiving channel stored in the sensitivity map storage unit 31 with the second index calculated using the reference image of each channel is used. It is the same as the first embodiment to correct the sensitivity distribution of the reference coil and use it for the sensitivity correction. The present embodiment is characterized by a method of acquiring the sensitivity distribution (sensitivity map stored in the sensitivity map storage unit) of the multi-channel receiving coil, which is the premise thereof.

The method for acquiring the sensitivity distribution is roughly divided into a method using a phantom and a method using a simulation, and any method may be adopted. In this embodiment, a method of creating a sensitivity map by simulation and correcting it with a sensitivity map obtained by phantom imaging will be described.

FIG. 8 shows a functional block diagram of the signal processing unit 7 in the present embodiment, mainly the sensitivity calculation unit. As shown in the figure, the sensitivity calculation unit 30 includes the sensitivity map storage unit 31, the sensitivity map coordinate conversion determination unit 32, and the sensitivity map creation unit 33, as well as the initial sensitivity map of the receiving coil (sensitivity map when the subject does not exist). It has a simulation unit 35 for calculating the above, and an initial sensitivity map adjusting unit 36 for matching the initial sensitivity map calculated by the simulation unit 35 with the image obtained by phantom imaging.

Hereinafter, the processing of the signal processing unit 7 of the present embodiment will be described with reference to FIG.
First, the simulation unit 35 calculates the sensitivity map by a known simulation technique such as the FDTD method (S801). In the simulation, the size of the phantom, the physical property value, the size of the target receiving coil, the value of the capacitor and inductor, the size between the elements, etc. are set, and the sensitivity distribution (sensitivity map) when a signal of magnetic resonance frequency is passed is set. calculate. In the present embodiment, as shown in FIG. 10, a phantom larger than the sensitivity map D (FIG. 10 (b)) obtained by the phantom can be set in the simulation so that a large subject can be handled. As a result, the initial sensitivity map S as shown in FIG. 10A is obtained.

Next, the initial sensitivity map adjusting unit 36 corrects the initial sensitivity map S, which is the simulation result, to bring it closer to the sensitivity map D obtained by actually imaging the phantom (S802). Here, in the phantom imaging, only the reception sensitivity can be extracted by removing the influence of the transmission non-uniformity by using an adibatic pulse. The B1 map may be obtained to remove the influence of non-uniform transmission. Further, a surface coil may be placed to increase the reception sensitivity including the influence of the surface coil.

As a method of bringing the initial sensitivity map S closer to the sensitivity map D of the phantom imaging, specifically, coordinate transformation (rotation, translation, scaling) by the equation (5) for the initial sensitivity map of each channel and each channel i By repeating the multiplication of different constants Ki, the coordinate conversion formula and the constant that minimize the difference between S'i (r) and D (r) are calculated. Here, the difference between the initial sensitivity map and the phantom image is compared for the region where the subject (phantom) exists.
[Number 5]
S'i (r) = Ki x Si (f (r)) (5)
In the formula, f represents the coordinate transformation formula f (r → r').

As the minimization method, a known method such as a method using the average of the squares of the differences between the initial sensitivity map obtained by simulation and the phantom image can be used, but here it is not necessary to exactly match the two. A rough correction is fine.

Finally, the phantom image D is divided by the roughly modified simulation sensitivity map S'(r) to obtain the coefficient (coefficient for each pixel) β, and the coefficient β having this distribution is the modified simulation sensitivity S'(r). (S803). At this time, in order to smooth the distribution of the coefficient β, it is smoothed with a median filter or a low-pass filter. Further, for the pixels in the region where the subject (phantom) does not exist in the phantom captured image D, the coefficient β calculated in the region where the subject exists is extrapolated (S804).
A known technique can be used for extrapolation, but since the coefficient β has a property of being constantly 1 if the simulation is complete, extrapolation can be performed stably by a simple process. For example, the process of replacing with the coefficient of the point at the position closest to the point to be extrapolated may be used.

By the above processing, the correction of the initial sensitivity map obtained by the simulation is completed. The modified sensitivity map (FIG. 10 (c)) is used to create the reception sensitivity map (sensitivity map for the subject) described in the first embodiment.

In the MRI apparatus of the present embodiment, the sensitivity maps of a plurality of receiving channels stored in the storage unit (sensitivity map storage unit) are sensitivity maps obtained by correcting the sensitivity image obtained in the simulation with a phantom image. According to this embodiment, a sensitivity map other than the subject used for creating the sensitivity map can be created by freely setting the phantom size, physical property values, and the like.

In the above description, in the process S802 for modifying the initial sensitivity map so as to roughly approach the phantom imaging result, the same coordinate transformation (rotation, translation, scaling) for each channel and multiplication of different constants for each channel are performed. I went, but it is not limited to this method. For example, you may only multiply the constants that are different for each channel, or you may omit the multiplication of the constants.

In the above explanation, a large phantom was set in the simulation and corrected to match the phantom imaging result, but from the same phantom simulation result as the phantom imaging, correction (rotation, scaling, translation, amplification, coefficient) Then, the correction may be applied to the simulation result of a large phantom to create a sensitivity map (sensitivity map to be stored in the sensitivity map storage unit).

Further, in the above explanation, the case of creating one sensitivity map (sensitivity map to be stored in the sensitivity map storage unit) has been described, but the conditions are such that they can be used properly according to the conditions (position, size, etc.) of the subject. Multiple sensitivity maps with different characteristics may be prepared in advance.

<Third Embodiment>
The feature of this embodiment is that a sensitivity map including phase information is created as a sensitivity map of the receiving coil for imaging. By obtaining the sensitivity map including the phase information, it is possible not only to correct the sensitivity unevenness (so-called sensitivity correction) but also to use the receiving coil sensitivity map for parallel imaging.

In the present embodiment, the sensitivity map storage unit 31 stores the sensitivity map of each channel of the receiving coil of a plurality of channels as a complex number. For example, the signal obtained by the receiving coil by imaging a homogeneous phantom is complex data including signal strength and phase, but when only sensitivity unevenness is corrected, only the signal strength, which is the absolute value of sensitivity, is available. The signal strength is the sensitivity map. However, the present embodiment also makes it possible to apply to parallel imaging or the like using reception sensitivity information including phase in, for example, image reconstruction calculation by using phase information.

The receiving coil for which the sensitivity map creating unit of the present embodiment creates a receiving sensitivity map may be a multi-channel coil for correcting the sensitivity map of another surface coil for imaging, but is preferably itself. It is a multi-channel coil for imaging. In the following description, a case where the multi-channel coil for imaging is used will be described. As shown in FIG. 11, the configuration of the signal processing unit 7 other than the sensitivity map storage unit 31 is the absolute value / phase correction unit for handling complex numbers in the configuration of the functional block diagram illustrated in FIG. 3 in the first embodiment. (Phase correction unit) 37 is added. Hereinafter, the operation of the signal processing unit 7 of the present embodiment will be described with reference to FIG. 12, focusing on the points different from those of the first embodiment.

First, in the coordinate conversion determination process S501, the absolute value (absolute value of each pixel) of the reference image obtained in each channel of the reference coil by the reference scan is set as an n-dimensional vector, and each is stored in the sensitivity map storage unit 31. Let the absolute value of the sensitivity distribution of the channel (absolute value of each pixel) be an n-dimensional vector. A coordinate transformation formula for coordinate transformation of the sensitivity map is calculated so that the angle formed by the vector of the reference image and the vector of the sensitivity map becomes smaller as a whole. This technique is similar to S403 of the first embodiment and includes scaling, translation, and rotation.

Next, using the coordinate conversion formula obtained in S501, the sensitivity map (complex number) of each channel stored in the sensitivity map storage unit 31 is coordinate-converted (S502). Next, the absolute value and the phase value of the sensitivity map after the coordinate conversion are corrected (S503). Therefore, first, for each pixel, the coefficient ki that satisfies the equation (6) is calculated from the absolute value of the sensitivity distribution and the absolute value of the reference image.

[Number 6]
｜ Ri ｜ ＝ ki × ｜ Si ｜ (6)
In the equation, Si (i indicates that it is the i-th channel, and an integer having the maximum number of channels) is the sensitivity map of the coordinate-transformed channel i, and Ri is the reference image of the channel i. Using this coefficient ki, the sensitivity map (complex number) R'i of each channel is calculated from the following equation (7).

[Number 7]
R'i = ki x Si x exp (j x di) x (| S | / | R |) (7)
In the equation, S is a vector having the pixel value of the sensitivity map of each channel as an element (n-dimensional vector when the number of channels is n), and R is a vector having the pixel value of the reference image as an element. j is an imaginary unit, and di is the difference (phase difference) between the argument (arg (Ri)) of the sensitivity map Ri and the argument arg (Si) of the reference image Si, and is defined by the equation (8).

[Number 8]
di = arg (Ri) -arg (Si) (8)
Both "arg (Ri)" and "arg (Si)" represent the declination after the unwrapping process.
In the sensitivity map R'i of each channel obtained by the equation (7), the ratio between the channels is the same as the ratio between the channels in the reference image, and the sensitivity map of the entire reference coil obtained by synthesizing the sensitivity map Ri is obtained. Will be flat.

If the reference image of each channel has a lower resolution than the sensitivity map for each channel stored in the sensitivity map storage unit 31, or if there is a region where an accurate reference image cannot be obtained due to the presence of metal or the like, those For the missing region or the missing pixel, ki and di are obtained by interpolation, respectively, and then R'i is obtained from the equation (7). As a result, it is possible to obtain a high-resolution sensitivity map from a low-resolution reference image, or to obtain a sensitivity map of a region where an accurate reference image cannot be obtained.

The sensitivity map of the multi-channel receiving coil whose absolute value and phase are determined in this way is a known parallel imaging using the NMR signal received in each channel of the multi-channel receiving coil if the main imaging of the parallel imaging is performed. It can be used for an operation based on the above, for example, a folding expansion operation using k-space data and a folding removal operation using real space data (S504).

If this imaging uses a surface coil different from the multi-channel receiving coil for which the sensitivity map was calculated by the above process, the sensitivity distribution of this surface coil is corrected as in the first embodiment. , Sensitivity correction may be performed on the image obtained by the surface coil.

In the MRI apparatus of the present embodiment, the sensitivity maps of a plurality of receiving channels stored in the storage unit (sensitivity map storage unit) and the plurality of reference images have complex pixel values, and the sensitivity map creation unit has a plurality of sensitivity maps. Further includes a sensitivity map of the receiving channel of the above and a phase correction unit for aligning the phases of the plurality of reference images. Further, the calculation unit (sensitivity calculation unit) can create an image by performing a calculation based on parallel imaging using the reception sensitivity map of each reception channel whose phase has been corrected by the sensitivity map creation unit.

According to this embodiment, the sensitivity map of each channel is treated as a complex number, the coordinates are converted so that the relationship between the channels becomes the same as the relationship between the channels of the image (reference image) when the subject is present, and the absolute value is obtained. And since the sensitivity information having the phase is obtained, accurate sensitivity correction is possible, and parallel imaging using the sensitivity information is also possible.

Further, according to the present embodiment, when creating a sensitivity map, it is possible to obtain a high-resolution sensitivity map even for a reference image having low resolution or lack of information. Therefore, it is possible to shorten the imaging time of the reference scan.

<Fourth Embodiment>
In the first to third embodiments, the sensitivity map of the receiving coil is created by using the coordinate conversion formula derived from the sensitivity map acquired in advance for each channel and the reference image for each channel for the receiving coil having a plurality of channels ( In this embodiment, a sensitivity map of a receiving coil (second receiving coil) different from the receiving coil (first receiving coil) having a plurality of channels from which the coordinate conversion formula is derived is obtained. , Created using that coordinate transformation.

The combination of the first receiving coil and the second receiving coil is not particularly limited as long as the first receiving coil is a receiving coil having a plurality of channels and the sensitivity distributions of both receiving coils overlap, for example. , As shown in FIG. 13, there are cases where a pair of butterfly coil (2-channel coil) 61 and a solenoid coil 62 are combined, or a CRC coil 63 is further added to the first receiving coil. Further, in the present embodiment, the sensitivity map of the first receiving coil, the sensitivity map of each channel of the first receiving coil, and the sensitivity map of the second receiving coil are stored in advance in the sensitivity map storage unit 31. And. The method of acquiring the sensitivity map stored in advance in the sensitivity map storage unit 31 is not particularly limited. For example, an acquisition method such as phantom imaging, simulation, or correction of what is calculated by simulation as in the second embodiment with the phantom imaging result can be adopted.

Hereinafter, the processing of the signal processing unit of the present embodiment will be described by taking as an example the receiving coil 60 which is a combination of the 2-channel receiving coil 61 and the solenoid coil 62 as shown in FIG. FIG. 14 shows the processing flow.

First, a reference scan is performed using the 2-channel receiving coil 61, and reference images of Ch1 and Ch2 of each channel are acquired (S701). A two-dimensional vector R whose element is the pixel value of the corresponding pixel of the reference image of each channel is set, and its declination θ is obtained. On the other hand, for the sensitivity distribution of each channel Ch1 and Ch2 obtained in advance for the 2-channel receiving coil 61, a two-dimensional vector S having the pixel value of the corresponding pixel as an element is set, and the declination φ thereof is obtained. The declination θ and φ are calculated for all the pixels or the pixels in the preset region, respectively, and two declination distributions are obtained (S702). Similar to the first embodiment, the coordinate conversion formula is determined so that the declination distribution obtained from the sensitivity map has a distribution similar to the declination distribution obtained from the reference image (S703).

Using this coordinate conversion formula, the sensitivity distribution obtained in advance for the solenoid coil 62 is coordinate-converted to obtain a sensitivity map of the solenoid coil 62 (S704). As a result, a sensitivity map for the subject can be obtained, so that the accuracy of sensitivity correction of the image captured by the solenoid coil 62 can be improved.

In the above example, the sensitivity map of the solenoid coil 62 is created, but it goes without saying that the sensitivity maps of each channel Ch1 and Ch2 of the 2-channel receiving coil 61 may be created by using the coordinate conversion formula. ..

Furthermore, in the above example, the second is using the sensitivity map (sensitivity map to be stored in the sensitivity map storage unit) of the first receiving coil (2-channel receiving coil 61) and the coordinate conversion formula created based on the reference image. Although the sensitivity map (sensitivity map for the subject) of the receiving coil (solenoid coil) has been created, the first receiving coil and the second receiving coil may partially overlap. For example, in the above-mentioned example of the combination of the 2-channel receiving coil 61 and the solenoid coil 62, a vector whose elements are the pixel values of the channel of the receiving coil 61 and the sensitivity map of the solenoid coil (sensitivity map to be stored in the sensitivity map storage unit). For the vector whose elements are the pixel values of the reference image and the reference image, a coordinate conversion formula is calculated so that their distributions (or distributions of deviation angles) are equal, and the sensitivity map (sensitivity map) of the solenoid coil is used using this coordinate conversion formula. The sensitivity map to be stored in the storage unit) may be converted to create a sensitivity map for the subject.

According to the present embodiment, in a receiving coil composed of a plurality of coils, when there is an element that serves as a reference for sensitivity correction such as a solenoid coil, it is possible to correct the reference sensitivity unevenness and perform efficient sensitivity correction. It can be carried out.

Although the embodiments of the present invention have been described above, the present invention is not limited to the individual embodiments. For example, unless technically inconsistent, the configurations, procedures, or modifications of the respective embodiments are described in other embodiments. It is also possible to combine with or omit a configuration that is not essential in the present invention.

2: Static magnetic field generator 3: Diagonal magnetic field generator 4: Sequencer 5: Transmitter, 6: Receiver, 7: Signal processing unit, 8: Central processing device (CPU), 14a: Transmit coil, 14b: Receive coil, 18: External storage device, 19: Display, 20: Operation unit, 30: Sensitivity calculation unit, 31: Sensitivity map storage unit (storage unit), 32: Sensitivity map coordinate conversion determination unit, 321: Index calculation unit, 322: Coordinate conversion formula calculation unit, 33: Sensitivity map creation unit, 35: Simulation unit, 36: Initial sensitivity map adjustment unit, 37: Absolute value / phase correction unit, 40: Image creation unit, 41: Sensitivity correction unit, 60 : Receiving coil, 61: 2 channel receiving coil, 62: Solenoid coil, 63: CRC coil.

Claims (13)

An imaging unit having a receiving coil having a plurality of receiving channels, receiving a nuclear magnetic resonance signal to be inspected, and acquiring an image to be inspected.
A calculation unit that performs calculations necessary for creating an image of the inspection target using the sensitivity information of the receiving coil, and
It is a magnetic resonance imaging device equipped with
A storage unit that stores the sensitivity map of the receiving coil acquired in advance other than when the inspection target is set , and the sensitivity map of the plurality of receiving channels or the map derived from the sensitivity maps of the plurality of receiving channels is further stored. Prepare,
The calculation unit
A first index map showing the relationship between the sensitivity maps of the plurality of receiving channels and a second index map showing the relationship between the plurality of reference images acquired by the imaging unit for each receiving channel of the receiving coil for the inspection target. A coordinate conversion formula calculation unit that calculates a coordinate conversion formula that transforms the coordinates of the first index map so as to minimize the difference from the index map.
A magnetic resonance imaging apparatus comprising: a sensitivity map creating unit that creates a reception sensitivity map when the inspection target exists by using the coordinate conversion formula. The magnetic resonance imaging apparatus according to claim 1.
The first index map is a map in which any one of the pixel value ratio of the sensitivity map of each receiving channel, the vector having the pixel value as an element, and the declination angle of the vector is used as the pixel value. ,
The second index map is a map in which any one of the pixel value ratio of the reference image of each receiving channel, the vector having the pixel value as an element, and the declination angle of the vector is used as the pixel value. A magnetic resonance imaging apparatus characterized in that. The magnetic resonance imaging apparatus according to claim 1.
The calculation unit is a magnetic resonance imaging apparatus including a sensitivity correction unit that corrects an image to be inspected by using a reception sensitivity map created by the sensitivity map creation unit. The magnetic resonance imaging apparatus according to claim 1.
The sensitivity map creating unit is a magnetic resonance imaging apparatus characterized in that the receiving sensitivity map as a whole is created by combining a plurality of channels for a receiving coil having the plurality of receiving channels. The magnetic resonance imaging apparatus according to claim 1.
The receiving coil having the plurality of receiving channels is a wide area receiving coil.
The calculation unit uses the reception sensitivity map of the wide area receiving coil created by the sensitivity map creation unit and the sensitivity map obtained in advance for an imaging receiving coil different from the wide area receiving coil. A magnetic resonance imaging apparatus further comprising a sensitivity map correction unit that corrects a sensitivity map of a receiving coil for imaging. The magnetic resonance imaging apparatus according to claim 1.
The first index map or the sensitivity map of the plurality of receiving channels stored in the storage unit is the phantom image obtained by imaging the phantom using the receiving coil having the plurality of receiving channels and the sensitivity obtained by the simulation. A magnetic resonance imaging apparatus characterized by being a first index map or sensitivity map created using at least one of the images. The magnetic resonance imaging apparatus according to claim 6.
A magnetic resonance imaging apparatus characterized in that a first index map or a sensitivity map of a plurality of receiving channels stored in the storage unit is created by correcting a sensitivity image obtained in the simulation with the phantom image. The magnetic resonance imaging apparatus according to claim 1.
The sensitivity map stored in the storage unit is a sensitivity map of the plurality of receiving channels, and the sensitivity maps of the plurality of receiving channels and the plurality of reference images have complex pixel values.
The sensitivity map creating unit is a magnetic resonance imaging apparatus further comprising a sensitivity map of the plurality of receiving channels and a phase correction unit for aligning the phases of the plurality of reference images. The magnetic resonance imaging apparatus according to claim 8.
The calculation unit is a magnetic resonance imaging apparatus characterized in that an image is created by performing an operation based on parallel imaging using a reception sensitivity map of each reception channel whose phase has been corrected by the sensitivity map creation unit. .. The magnetic resonance imaging apparatus according to claim 1.
The sensitivity map creation unit creates the reception sensitivity map for a part of the reception channels of the reception coil having the plurality of reception channels, or for another coil whose sensitivity region overlaps with the reception coil having the plurality of reception channels. A magnetic resonance imaging device characterized by being created. The magnetic resonance imaging apparatus according to claim 10.
The receiving coil having the plurality of receiving channels is a multi-channel coil having sensitivity in a region where one receiving channel is a solenoid coil and the other receiving channel overlaps the solenoid coil.
The coordinate conversion formula calculation unit creates a coordinate conversion formula for at least the plurality of channel coils by using the first index map and the second index map.
The sensitivity map creating unit is a magnetic resonance imaging apparatus characterized in that a receiving sensitivity map of the solenoid coil is created by using the coordinate conversion formula. It is a method of calculating the reception sensitivity map of the receiving coil with respect to the inspection target by using the sensitivity map of the receiving coil acquired except when the inspection target is set .
For a receiving coil having a plurality of receiving channels, the relationship between the pixels of the sensitivity maps of the plurality of receiving channels acquired except when the inspection target is set is calculated as the first index, and the first index map is obtained. ,
The relationship between the pixels of the reference images of the plurality of reception channels acquired for the inspection target is calculated as the second index, and the second index map is obtained.
The coordinate conversion formula derived so that the first index map and the second index map match is applied to the sensitivity map of the receiving coil acquired except when the inspection target is set, and the receiving sensitivity map is calculated. A method for calculating a sensitivity map of a receiving coil, which comprises the above. The method for calculating the reception sensitivity map of the reception coil according to claim 12.
The sensitivity map of the receiving coil is characterized in that the coordinate conversion formula is applied to a receiving coil different from the plurality of receiving channels, and the receiving sensitivity map is created for the receiving coil different from the plurality of receiving channels. Calculation method.

Images