JP4306470B2 - Magnetic resonance imaging system - Google Patents

Description

  The present invention relates to a magnetic resonance imaging (hereinafter abbreviated as MRI) apparatus for displaying an arbitrary cross section of a subject by utilizing a nuclear magnetic resonance (hereinafter abbreviated as NMR) phenomenon, and in particular, phase encoding by an RF receiving coil. The relative position between the subject and the RF receiver coil using table movement in an imaging method that unfolds aliasing artifacts that occur in an image reconstructed from echo signals acquired by thinning out the image using the sensitivity distribution of the RF receiver coil The present invention relates to an MRI apparatus that can obtain an artifact-free 3D image by acquiring an echo signal while changing.

In the MRI apparatus, the sequence is repeatedly executed while changing the phase encoding amount, and an echo signal necessary for reconstruction of one image is acquired. Therefore, the number of repetitions greatly affects the image shooting time.
Therefore, as one of the techniques for shortening the imaging time by reducing the number of phase encodings, a high-speed imaging method using a plurality of RF receiving coils has been proposed. In this method, the number of repetitions is reduced by thinning out the phase encoding and measuring the echo signal. Usually, if measurement is performed with phase encoding thinned out, aliasing artifacts occur in the reconstructed image, but this method expands the image by performing matrix calculation using the sensitivity distribution of each RF receiving coil, and then folding it back. Remove artifacts. Thereby, in general, the imaging time can be shortened by the number of RF receiving coils used for imaging. Such an image reconstruction method is called parallel MRI, and its details are disclosed in [Patent Document 1] and [Patent Document 2].

  Usually, as an example of an RF receiving coil used for parallel MRI, a technique called “multiple RF coil” or “phased array coil” using a plurality of receiving coils is used. Multiple RF coils are a combination of multiple relatively high-sensitivity small RF receiver coils, and by synthesizing the signals acquired by each coil, the field of view is expanded while maintaining high sensitivity of the RF receiver coil, and high sensitivity This is a dedicated RF coil for reception.

Examples of multiple RF coils for horizontal magnetic field heads are [Non-patent Document 1], and examples of QD multiple RF coils for horizontal magnetic field heads are [Non-Patent Document 2] and [Non-Patent Document 3]. There is. Each of these is constructed by adhering a small surface coil adjacent to the head surface.
There are [Non-patent document 4] and [Non-patent document 5] as the multiple RF coils for the vertical magnetic field head and neck. In these, solenoid coils are arranged at a certain distance.
As the QD multiple RF coil for the horizontal magnetic field abdomen, there are [Non-patent document 6] and [Non-patent document 7]. These are constructed by adhering small surface coils adjacent to the abdomen and back.
That is, the parallel MRI proposed so far uses a plurality of coils, and removes the folding artifacts using the difference in sensitivity distribution of each coil.

“Array Head Coil for Improved Functional MRI” (Christoph Leussler), 1996 ISMRM abstract p.249 “Helmet and Cylindrical Shaped CP Array Coils for Brain Imaging: A Comparison of Signal-to-Noise Characteristics” (H.A.Stark, E.M.Haacke), 1996 ISMRM abstract p.1412 “8-element QD domed head array coil using inductive decoupler” (Tetsuhiko Takahashi, et al), 1998 ISMRM abstract p.2028 “Head-neck quadrature multiple RF coil for vertical magnetic field MRI” (Tetsuhiko Takahashi, Yoshikuni Matsunaga), 1997 ISMRM abstract p.1521 "Multi-RF coil enhances wide-field sensitivity of head and neck MRI" (Tetsuhiko Takahashi, Yoshikuni Matsunaga), Medical Imaging Technology, vol.15, no.6, pp.734-741 (1997) “Four Channel Wrap-Around Coil with Inductive DecoupleRFor 1.5T Body Imaging” (T. Takahashi et al), 1995 ISMRM abstract p.1418 “High-sensitivity wraparound RF coil for MRI-Application of induction decoupler to multiple RF coil-” IEICE Transactions, vol.J80-D-II, no.7, pp.1964-1971 (1997)

  On the other hand, a technique for imaging the whole body using high-speed imaging of MRI is known. In this method, the whole body of the subject is divided into a plurality of regions, each region is imaged, the table is moved to the next region, and the process of imaging the region is repeated to image the whole body of the subject. It is called multi-station MRI.

  In particular, a technique for imaging the blood vessels of the whole body of the subject using the multi-station MRI technique is called multi-station MRA, and for example, an artery of the whole body is imaged while injecting a Gd contrast agent into the blood vessels. In multi-station MRA, the trunk, pelvis, and lower limbs are each covered with a dedicated RF receiver coil, and the descending aorta from the aortic arch of the subject descends to the abdominal aorta and femoral artery, and the contrast agent is a peripheral blood vessel, As the bolus flows to the lower knee, the table is moved twice in about 15 seconds, the thigh in about 15 seconds, and the lower leg in about 15 seconds. Field of view (FOV), i.e., appropriate static magnetic field area, appropriate gradient magnetic field generation area, appropriate RF transmission area and appropriate RF reception area. In 3.5 s to 5 s), the table (and subject) is moved and the next part is imaged (Non-patent Document 8).

In addition, an example of capturing a whole body structure image instead of angiography has been reported (Non-patent Document 9), and a parallel MRI technique using a moving table is disclosed for two-dimensional imaging (No. 1 of Patent Document 1). 7 examples).
Imaging by MR angiography, Masaki Ishida et al., Imaging, vol. 23, no. 8, pp901-9102003 M. Bock et al. Whole-body MRI: A simple approach using automatic table movement and dedicated post-processing, Proceeding of International Society for Magnetic Resonance in Medicine, 2002), (Yodong Zhu et al. Extended field-of-view imaging with table translation and frequency sweeping, Magnetic Resonance in Medicine 49: 1106-1112,2003), (DG.Kruger et al. Continuously moving table method for extended FOV 3D MRI, Proceeding of International Society for Magnetic Resonance in Medicine, 2002 JP 2002-10992 A

  In the multi-station MRI, as described above, generally, a plurality of areas of a subject are imaged using a dedicated RF receiving coil suitable for imaging the area. In this case, due to the non-uniformity of the sensitivity of these RF receiving coils, there arises a problem in that the brightness of the image is different at the connecting portion when images taken at each station are connected.

  More specifically, in multi-station MRI, if there is a sensitivity difference between the RF receiving coil and the adjacent RF receiving coil, the brightness of the image becomes uneven near the connection when each image is connected and displayed. There is. This problem becomes prominent when 3D shooting is performed at each station. That is, in the case of 3D imaging, the sensitivity unevenness of each RF receiving coil is generated three-dimensionally, so that the luminance unevenness between the adjacent station images to be connected differs for each slice. Therefore, adjustment for each slice is required when connecting the images, which is very troublesome and troublesome for the operator.

  In general, 3D shooting requires a longer shooting time than 2D shooting. Therefore, especially in the case of a multi-station MRA that uses a contrast medium, the contrast medium will flow completely if each part (station) is not imaged quickly. Movement is desired.

  In summary, in multi-station MRI that repeats 3D imaging while moving the table, there is a problem of reducing the luminance unevenness of the reconstructed image caused by the RF receiving coil or the like without substantially extending the imaging time.

  Therefore, the present invention has been made to solve the above-described problems in 3D multi-station MRI, and it is an object of the present invention to reduce uneven brightness of the screen when connecting the images of each station in 3D multi-station MRI. To do. In addition, it aims to increase the speed of 3D shooting at each station and reduce the overall shooting time.

In order to solve the above problems, the present invention is configured as follows. That is,
A static magnetic field generating means for generating a uniform static magnetic field space in which the subject is arranged;
An RF receiver coil for receiving a nuclear magnetic resonance signal from the subject;
A table on which the subject is placed and moved;
Measuring means for thinning out at least phase encoding and measuring the nuclear magnetic resonance signal to obtain thinned-out k-space data ;
Image reconstruction means for acquiring an image without aliasing using the thinned-out k-space data and the reception sensitivity distribution data of the RF receiving coil;
Control means for controlling execution of the table movement, the thinning measurement, and the image reconstruction;
In a magnetic resonance imaging apparatus comprising:
The RF receiving coil is one and is disposed in a position in the uniform static magnetic field space,
The measurement means obtains first thinned-out k-space data by performing thinning measurement when a desired part of the subject is at a first relative position with respect to the one RF receiving coil, When the relative position is obtained, the thinning measurement is performed to obtain the second thinning k-space data,
The image reconstruction unit is configured to generate an unfolded image related to the desired portion from the first thinned-out k-space data, the second thinned-out k-space data, and the sensitivity distribution data of the one RF receiving coil. get.

  According to this embodiment, only one RF receiving coil is used, and the reception sensitivity of the RF receiving coil as viewed from the subject, which is virtually generated based on the relative positions of the subject and the RF receiving coil being different. A parallel MRI that unfolds a fold generated on an image can be applied using the difference in distribution. For this reason, in the 3D multi-station MRI, the luminance unevenness on the composite image can be reduced without substantially extending the imaging time.

In a preferred second embodiment of the present invention, in the MRI apparatus according to the first embodiment, the measuring means measures a three-dimensional nuclear magnetic resonance signal by applying slice encoding to the first and first 2 at the relative positions of the first and second thinned three-dimensional k-space data,
The image reconstruction unit is configured to return the desired portion from the first thinned-out three-dimensional k-space data, the second thinned-out three-dimensional k-space data, and the sensitivity distribution data of the one RF receiving coil. A three-dimensional image with no image is acquired.
Further, in a preferred third embodiment of the present invention, in the MRI apparatus of the second embodiment, the measurement means performs the thinning measurement by moving the table at least every step of the slice encoding,
The image reconstruction means aligns the coordinates of the movement direction of the subject and performs Fourier transform in the slice encoding direction to obtain the unfolded three-dimensional image.
According to these embodiments, it is possible to reduce the dead time of shooting with the table moving. For this reason, in the 3D multi-station MRI, the entire imaging time can be shortened, and the burden on the subject can be reduced.

  As described above, according to the present invention, in the 3D multi-station MRI, the luminance unevenness on the connected image based on the sensitivity difference between the conventional RF receiving coil and the adjacent RF receiving coil is not substantially extended. Can be reduced. This eliminates the need for the operator to adjust the luminance unevenness between the images when connecting the images of the stations, and reduces the burden on both subjects because the entire imaging time is reduced for the subject. Will be.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment of the invention, and the repetitive description thereof is omitted.
First, a block diagram of an embodiment of an MRI apparatus to which the present invention is applied is shown in FIG. The MRI apparatus in FIG. 7 includes a magnet 402, a gradient magnetic field coil 403, an RF transmission coil 404, and an RF reception coil 405.
The magnet 402 generates a uniform static magnetic field in the body axis direction (horizontal magnetic field method) or in the direction perpendicular to the body axis (vertical magnetic field method) in the space around the subject 401. The normal conduction method and the superconductivity method Alternatively, a permanent magnet type static magnetic field generation source is arranged around the subject 401 to generate a static magnetic field.

  The gradient magnetic field coil 403 is a coil that independently generates a gradient magnetic field in each of the three axis directions of X, Y, and Z. From the gradient magnetic field power source 409 that drives each coil according to control from the control unit 411 described later. In response to the supply of current, gradient magnetic fields Gx, Gy, and Gz in the X-axis, Y-axis, and Z-axis directions are applied to the subject 401. More specifically, a slice direction gradient magnetic field pulse (Gs) is applied in one of X, Y, and Z to set a slice plane for the subject 401, and the phase encode direction gradient magnetic field is applied to the remaining two directions. A pulse (Gp) and a frequency encoding direction gradient magnetic field pulse (Gf) are applied, and position information in each direction is encoded in the echo signal.

  The control unit 411 is a control unit that repeatedly applies a high-frequency magnetic field pulse (hereinafter abbreviated as an RF pulse) and a gradient magnetic field pulse in a predetermined pulse sequence, and various commands necessary for collecting tomographic image data of the subject 401. Is sent to the gradient magnetic field power source 409, the RF transmitter 410, and the signal detector 406. The control unit 411 of the MRI apparatus to which the present invention is applied further controls the movement of the table in conjunction with the pulse sequence.

  The RF transmission unit 410 irradiates an RF pulse to induce an NMR phenomenon on the atomic nuclear spin constituting the biological tissue of the subject 401, and outputs from the RF transmission unit 410 at a timing according to a command from the control unit 411. The RF pulse is irradiated to the subject 401 by supplying the high frequency pulse to the RF transmission coil 404 disposed in the vicinity of the subject 401.

  The signal detection unit 406 detects an echo signal (NMR signal) emitted by the NMR phenomenon of the nuclear spin constituting the living tissue of the subject 401, and is induced by an RF pulse emitted from the RF transmission coil 404. An echo signal in response to the subject 401 is detected by the RF receiving coil 405 disposed close to the subject 401.

  The signal processing unit 407 performs processing such as signal processing and image reconstruction on the echo signal from the signal detection unit 406, outputs the resulting image of the subject 401 to the display unit 408, The data is output to an external storage device (not shown) such as a magneto-optical disk.

  The table 412 is for the subject to lie down. Under the control of the control unit 411, the table 412 can be moved in the parietal-foot (HF) direction of the subject 401 (arrow 413 in the figure), and the position of the subject 401 is adjusted while matching with the execution of the pulse sequence. Move in HF direction. In the MRI apparatus to which the present invention is applied, the position of the RF receiving coil 405 is fixed in the imaging space and does not move in the HF direction together with the table, and the subject 401 moves through the inside of the RF receiving coil 405.

  Currently, MRI imaging targets are protons, which are the main constituents of specimens, as they are widely used in clinical practice. By imaging the spatial distribution of proton density and the relaxation phenomenon of excited states, the functions of the human head, abdomen, limbs, etc. Take a dimensional shot.

  Next, the flow of image reconstruction from echo signal detection in the above-described MRI apparatus will be described. FIG. 6 shows an example. In FIG. 6, an RF receiving coil 405 includes a coil unit 301 and a preamplifier 302, and an echo signal detected by the coil unit 301 is amplified by the preamplifier 302 and output. The preamplifier 302 is preferably arranged as close to the coil unit 301 as possible. In the signal detection unit 406, one AD conversion / orthogonal detection circuit 303 is arranged in parallel, and the output from the preamplifier 302 is connected. In this signal detection unit 406, the amplified echo signal is converted into a digital quantity by AD conversion, divided into two orthogonal signals by orthogonal detection, and converted into complex echo data (hereinafter simply referred to as echo data). Is done. The signal processing unit 407 reconstructs an image of the subject 401 by Fourier transform (304) of the echo data. Furthermore, a calculation unit 305 that calculates the image data as necessary is provided. In the MRI apparatus to which the present invention is applied, the calculation unit 305 performs various calculations necessary for parallel MRI described later.

  Next, a 3D (three-dimensional) imaging method will be described. FIG. 9 shows an example of a general 3D gradient echo sequence as a typical pulse sequence used in this embodiment. In the pulse sequence shown in FIG. 9, 601 is a high frequency pulse, 602 is a slice selective gradient magnetic field pulse, 603 is a slice encode gradient magnetic field pulse, 604 is a phase encode gradient magnetic field pulse, 605 is a frequency encode gradient magnetic field pulse, 606 is an echo signal, Reference numeral 607 denotes an echo time TE, and reference numeral 608 denotes a repetition time (an interval of 601) TR.

  In 3D imaging, the slice encoding gradient magnetic field pulse 603 and the phase encoding gradient magnetic field pulse 604 are applied at different repetition times 608 (the area surrounded by the gradient magnetic field pulse waveform and the time axis) is changed to echo different slice encoding and phase encoding. This is applied to the signal 606, and the echo signal 606 obtained by each encoding is detected. This operation is repeated by the number of two encodings, and an echo signal necessary for reconstruction of a single 3D image is acquired at an image acquisition time determined by multiplying the repetition time 608 and the number of encodings. For the number of slice encoding and phase encoding, a combination of values such as 32, 64, 128, 256, 512, etc. is usually selected for one 3D image. Each echo signal is usually obtained as a time series signal composed of 128, 256, 512, and 1024 sampling data. These data are three-dimensionally Fourier transformed to reconstruct a 3D image.

  Next, parallel MRI will be described. In the present invention, a known parallel MRI technique is used in which the phase (or slice) encoding step interval is thinned out at a constant rate, thereby reducing the number of repetitions of the pulse sequence and shortening the imaging time. This decimation can be applied to either or both of slice encoding and phase encoding. In the following description, an example of applying decimation to phase encoding will be described. FIG. 5 shows a conceptual diagram of an example of parallel MRI that thins out phase encoding in 3D imaging.

  FIG. 5 (a) shows a case of normal imaging without thinning out phase encoding.Echo data group 202 (2021 to 2027) measured at each phase / slice encoding amount is represented in three-dimensional k-space (frequency encoding direction is Kx, The phase encoding direction is Ky and the slice encoding direction is Kz), and the echo data 201 for one 3D image is obtained. An image 206 of an arbitrary slice after Fourier transform is shown in FIG. 5 (c), and an entire image of, for example, a circular phantom is drawn on the screen (FOV) without generating a folding artifact.

  On the other hand, FIG. 5 (b) shows a case where the phase encoding step interval is doubled and the echo data measurement is thinned out. In this case, echo data 204 (2041 to 2043) is measured every other line in the phase encoding direction (Ky), and phase-encoded echo data corresponding to the position of 205 (2051 to 2054) is not measured. Since the amount of echo data measured in this way is halved, when the image is reconstructed by halving the matrix, the FOV in the phase encoding direction is halved as shown in Fig. 5 (d), and both ends cannot fit in the FOV. An image 208 is obtained in which the portion of is folded back to the opposite side. That is, this folding occurs when the upper side 2071 and the lower side 2072 of the normal image (FIG. 5C) overlap in the areas 2092 and 2091, respectively.

The aliasing on the image can be developed and removed by using known signal processing (matrix operation) disclosed in [Non-patent Document 13]. In this method, echo data with aliasing is obtained as reference data by measuring the reception sensitivity distribution of a plurality of RF reception coils with different reception sensitivity distributions in advance, and is measured for each RF reception coil by thinning out the phase encoding. Are expanded and removed by matrix calculation using the previously obtained reception sensitivity distribution for each RF receiving coil.
In high-speed imaging using a plurality of RF receiving coils having different reception sensitivity distributions, in principle, the number of phase encodings for each imaging can be reduced by the number N of coils used. That is, the number of phase encodes can be (1 / N times).

Here, an example of matrix operation for aliasing removal is shown below based on FIG. In FIG. 8 and the following description, for simplicity, a case will be described in which two element coils having different reception sensitivity distributions are used. First, it is known that the following relationship is established when a field of view is assigned to two RF receiving coils, FOV1 and FOV2 are set, and m1 and m2 are element coil numbers.
S ^{i, j1} = A ^{i, j} ₁₁ · m ^{i, j} ₁ + A ^{i, j} ₁₂ · m ^{i, j} ₂ (1)
S ^{i, j} ₂ = A ^{i, j} ₂₁ · m ^{i, j} ₁ + A ^{i, j} ₂₂ · m ^{i, j} ₂ (2)

Where S ^{i, j} ₁ is the image calculated from element coil 1, A ^{i, j} ₁₁ is the reception sensitivity distribution of element coil 1 in FOV1, m ^{i, j} ₁ is the magnetization distribution in FOV1, and A ^{i, j} ₁₂ is the reception sensitivity distribution of element coil 1 in FOV2, m ^{i, j} ₂ is the magnetization distribution in FOV2, S ^{i, j} ₂ is the image calculated from element coil 2, and A ^{i, j} ₂₁ is the element in FOV1 Sensitivity distribution of coil 2, A ^{i, j} ₂₂ is sensitivity distribution of element coil 2 in FOV2. Here, S, m, and A are matrices equal to the matrix size of FOV1 and FOV2. I and j are spatial coordinates. Therefore, the relative coil sensitivity distributions C ₁ and C ₂ for each field of view FOV are expressed as C ^{i, j} ₁ = A ^{i, j} ₂₁ / A ^{i, j} ₁₁ (3)
C ^{i, j} ₂ = A ^{i, j} ₁₂ / A ^{i, j} ₂₂ (4)
If defined, the following equations (5) and (6) are obtained from the above equations (1) to (4) by matrix calculation.

m ^{i, j} ₁ · A ^{i, j} ₁₁ = (S ^{i, j} 1−S ^{i, j} ₂ · C ^{i, j} ₂ ) / (1−C ^{i, j} ₁ · C ^{i, j} ₂ ) (5)
m ^{i, j} ₂ · A ^{i, j} ₂₂ = (S ^{i, j} 2 −S ^{i, j} ₁ · C ^{i, j} ₁ ) / (1−C ^{i, j} ₁ · C ^{i, j} ₂ ) (6)
The left side of the equations (5) and (6) is a magnetization distribution weighted by the reception sensitivity distribution of the element coil. By arranging the images of equations (5) and (6) side by side, an entire image without aliasing can be obtained.
The explanation of the above formulas (1) to (6) is shown in the case of 2 speeds with 2 coils for simplicity, but such a concept can be extended to 2 speeds 3 coils and 2 speeds 4 coils. it can.

  Next, a conventional multi-station MRI will be described with reference to FIGS. FIG. 3 (a) is a schematic diagram showing a conventional example in which the abdomen, thigh, and lower leg of a subject are imaged at three stations. The subject 101 is fixed on the table and moves in the y direction together with the table. It is assumed that the center of the static magnetic field of the MRI apparatus is on the line 106.

  Shooting takes place at 3 stations. That is, first, the abdomen 102 is photographed by photographing 1, the thigh 103 is photographed by photographing 2, and the lower leg 104 is photographed by photographing 3. At that time, in photographing the thigh 103 and the lower leg 104, the table 412 is moved before that to make the photographing FOV 121 coincide with the thigh 103 and the lower leg 104, respectively.

  The reception sensitivity distribution of the RF reception coil 405 is distributed around the center of the magnetic field. The position of the RF receiving coil 405 moves with the table, but the RF receiving coil 405 is installed in each imaging region. That is, the RF receiving coil 405-1 for the abdomen is used for imaging the abdomen 102, the RF receiving coil 405-2 for the thigh is used for imaging the thigh 103, and the lower leg 104 is used for imaging. An RF receiving coil 405-3 for the lower leg is used. FIG. 3B shows the reception sensitivity distributions (107-1 to 107-3) of these RF reception coils 405 group.

When the table is moved, the RF receiving coil 405 corresponding to the photographing FOV 121 is selected and photographed by switching with a switch (not shown). Note that the sensitivity range 122, subject position 123, and imaging FOV 121 in each imaging are as shown in FIG.
Alternatively, with one RF receiving coil 405, even if the table moves, it does not move with it, and is always fixed on the static magnetic field center 106 so that it is in a fixed position in the apparatus coordinate system. May be.

  In imaging 1, which is the first station, the abdomen 102 of the subject coincides with the imaging region of the MRI apparatus, and the abdomen 102 becomes the imaging FOV 121. Three-dimensional imaging of the abdomen 102 is performed at this position. The field of view is 30 cm and the thickness is 15 cm, for example. The reading direction is x, the phase encoding direction is y, and the slice encoding direction is z. The imaging matrix is 256 (x) x256 (y) x64 (z). When 256 (phase encoding) × 64 (slice encoding) encoded signals are acquired, image reconstruction at this position is performed. The reconstructed image is a coronal section (COR) image of the subject and the slice direction is AP (Anterior-Posterior), but the reception sensitivity distribution of the RF receiving coil 405-1 is also reflected in the image. A typical reception sensitivity distribution of the RF receiving coil 405-1 is 107-1 when plotted in the Y direction in a coordinate system (X, Y, Z) fixed to the subject. In this conventional example, for simplicity of explanation, it is assumed that the sensitivity distribution in the z direction is substantially uniform. The shooting time is TRx (number of phase encodes) x (number of slice encodes).

  Next, move the table 30 cm to the top of the y-direction and take a picture 2. In imaging 2, the subject's thigh 103 matches the imaging FOV 121 of the MRI apparatus. At this position, three-dimensional imaging of the thigh 103 is performed as described above. The reception sensitivity distribution of the RF receiving coil 405-2 reflected in the reconstructed image is like 107-2. Although not shown, the first image and the second image may be overlapped on the subject by, for example, about 3 cm by adjusting the FOV and the moving distance.

  Finally, the table is further moved 30 cm, and the imaging FOV 121 of the MRI apparatus is matched with the lower leg 104 of the subject, and imaging 3 is performed. At this position, three-dimensional imaging of the lower leg 104 is performed in the same manner as described above. The reception sensitivity distribution of the RF reception coil 405-3 reflected in the reconstructed image is like 107-3. The second image and the third image may overlap, for example, about 3 cm on the subject.

  The three images obtained as described above are connected in the Y direction by a coordinate system (X, Y, Z) fixed to the subject to obtain an image of the entire lower limb. As shown in the figure, the luminance distribution of the entire image is a series of three peaks, and the sensitivity is significantly reduced at the connected portion. Such uneven sensitivity causes shading (intensity unevenness) in the image. Such shading can be made uniform to some extent by a post process such as shading correction, but it does not essentially improve the signal-to-noise ratio of the signal degradation portion.

  FIG. 4 (a) is an example in which the abdomen, thigh, and lower leg of the subject are similarly imaged at 5 stations, and is a schematic diagram showing a method of reducing the sensitivity unevenness based on the conventional example. However, in Fig. 4 (a), there is only one RF receiving coil 405, and even if the table moves, it does not move along with it, and it is fixed so that it is always located on the static magnetic field center 106. An example is shown in a fixed position. FIG. 4B shows a substantial reception sensitivity distribution of the RF reception coil 405.

  In imaging 1, which is the first station, the abdomen 102 of the subject coincides with the imaging FOV 121 of the MRI apparatus. At this position, 3D imaging of the abdomen 102 is performed. The photographing FOV 121 is, for example, 30 cm and a thickness of 15 cm. The reading direction is x, the phase encoding direction is y, and the slice encoding direction is z. The imaging matrix is 256 (x) x256 (y) x64 (z). When 256 (phase encoding) × 64 (slice encoding) encoded signals are acquired, image reconstruction at this position is performed. The reconstructed image is a coronal section (COR) image of the subject and the slice direction is AP (Anterior-Posterior), but also reflects the reception sensitivity distribution of the RF receiving coil. A typical receiving sensitivity distribution of the RF receiving coil is 107-1.

Next, move the table to the top of the head 15 cm in the y direction and take a picture 2. In imaging 2, the imaging FOV 121 of the MRI apparatus coincides with a position between the abdomen 102 and the thigh 103 of the subject. At this position, three-dimensional imaging of this intermediate area is performed as described above. The reception sensitivity distribution of the RF reception coil reflected in the reconstructed image is as shown in 107-2.
Thereafter, the table is sequentially moved 15 cm, and imaging is repeated. In imaging 5, the lower leg 104 of the subject is matched with the imaging FOV 121 of the MRI apparatus. The reception sensitivity distribution of the RF receiving coil reflected in the reconstructed image thus obtained is as shown in 107-5. The sensitivity range 122, subject position 123, and imaging FOV 121 in each imaging are as shown in FIG.

The five images obtained as described above are connected in the Y direction in a coordinate system fixed to the subject to obtain an image of the entire lower limb. The luminance distribution of the entire image is such that five peaks are connected as shown in the figure, and the decrease in sensitivity at the connection portion is reduced. In this way, it is possible to return the sensitivity unevenness to be almost uniform, but with this method, the number of times of photographing is five, and the photographing time is extended to 5/3 times that in the case of FIG.
Note that the S / N ratio of the signal in the region where the sensitivity overlaps can be improved by making each signal value a root of a square sum for each pixel (square sum method).

Next, the present invention will be described. The present invention is a three-dimensional multi-station MRI that uses only one channel whole body RF receiving coil without using a plurality of RF receiving coils, and also moves the table by 1/2 of the imaging FOV. Take one shot. Thereby, the position of the subject corresponding to the imaging FOV does not change. However, since the position of the RF receiving coil is fixed in the MRI apparatus coordinate system, the sensitivity distribution of the RF receiving coil moves FOV / 2 in the moving direction of the table when viewed from the coordinates (subject coordinates) fixed to the subject. It will be. Accordingly, when viewed from the subject coordinates, the same imaging FOV is captured by two RF receiving coils having different reception sensitivity distributions in each imaging. Using this effect, the parallel MRI method is applied to each imaging to reduce the number of slice encoding or phase encoding, thereby shortening the imaging time and making the table movement amount small, thereby substantially reducing the RF receiving coil. Sensitivity unevenness is reduced, and unevenness in luminance when connecting each image to obtain a composite image is reduced.
Note that the reception sensitivity distribution used for the second imaging in the same imaging FOV can be obtained by shifting the reception sensitivity distribution used for the first imaging by FOV / 2 in the table moving direction, so that re-measurement is unnecessary.

Here, it will be described below that the aliasing removal in parallel MRI is possible due to the effect that the reception sensitivity distribution of one RF reception coil appears to be substantially different from the subject coordinates as the table moves. In (1) and (2), redefine each variable as follows.
S ^{i, j} ₁ : Image calculated from the reception sensitivity distribution 1 of the RF receiver coil when the table position is 1
m ^{i, j} ₁ : When the table position is 1, the magnetization distribution of the subject in FOV1
A ^{i, j} ₁₁ : When the table position is 1, the receiving sensitivity distribution 1 of the RF receiving coil in FOV1
A ^{i, j} ₁₂ : Reception sensitivity distribution 1 of RF receiving coil in FOV2 when table position is 1
S ^{i, j} ₂ : The image calculated from the reception sensitivity distribution 2 of the RF receiver coil when the table position is 2.
m ^{i, j} ₂ : When the table position is 2, the magnetization distribution of the subject in FOV2
A ^{i, j} ₂₁ : When the table position is 2, the receiving sensitivity distribution 2 of the RF receiving coil in FOV1
A ^{i, j} ₂₂ : When the table position is 2, the receiving sensitivity distribution 2 of the RF receiving coil in FOV2

  It can be understood that the present invention can be implemented because Equations (5) and (6) can be similarly derived by such replacement. In the above description of the present invention, the case where the photographing is performed twice for the same photographing FOV and the table is moved by the photographing FOV / 2 between photographings is shown, but N (≧ 3) times per photographing FOV. It is also possible to perform a parallel MRI by taking a picture and moving the table by the photographing FOV / N between the pictures. In this case, since the substantial sensitivity unevenness of the RF receiving coil can be further reduced, the brightness unevenness in the combined composite image can be further reduced. On the other hand, the shooting time is extended compared to the case where the shooting time is twice.

Next, the first embodiment of the present invention will be described. This first embodiment performs parallel MRI imaging at each station in three dimensions, and combines the images obtained at each station to obtain a composite image, reducing the above-described luminance unevenness without extending the imaging time. It is a form to do.
An example of the first embodiment will be described in detail with reference to FIG. In the embodiment of FIG. 1, when the abdomen, thigh, and lower leg of the subject are three-dimensionally imaged at three stations, as shown in FIG. This is an embodiment in which the entire imaging time is shortened to ½ by applying parallel MRI while moving the table one by one.

In this embodiment, a one-channel whole body RF receiving coil 405 is used in a vertical magnetic field type MRI apparatus. Typically, it is a one-turn solenoid coil (that is, a loop coil) having an opening in the head (H) -foot (F) direction (that is, the y direction) of the subject.
The direction of the static magnetic field is the vertical (z) direction and its strength is, for example, 0.7T. In this case, the resonance frequency of proton is about 29.8 MHz, and this high-frequency magnetic field is a rotating magnetic field on the xy plane.

  The RF receiving coil 405 is a solenoid coil designed to resonate in parallel at the proton resonance frequency, and detects the y-direction component of the rotating magnetic field. In general, it is known that the diameter of the RF receiving coil and the region in the y direction of sensitivity (for example, defined as a region including up to half the maximum sensitivity) are substantially equal. For example, if the diameter of the RF receiving coil 405 is 30 cm, the sensitivity region may be considered to be about 30 cm. Further, the maximum sensitivity exists on a plane including the RF receiving coil 405.

In the example of the MRI apparatus of FIG. 7 described above, in this embodiment, for example, the control of the movement of the table 412 and the imaging is performed by the control unit 411, and the image reconstruction process including the folding expansion process and the synthesis process of each station image are performed. The signal processing unit 407 performs reception sensitivity distribution of the RF receiving coil, measurement data acquired at each station, storage of each station image, and a combined composite image are stored in an external storage device.
In the present embodiment, unlike the conventional example of the multi-station MRI described above, the RF receiving coil 405 is unevenly distributed 15 cm from the magnetic field center to the top of the head.

  In imaging 1, the abdomen 102 of the subject coincides with the imaging FOV 121 of the MRI apparatus. At this position, 3D imaging of the abdomen 102 is performed. The photographing FOV 121 is, for example, 30 cm and a thickness of 15 cm. The reading direction is x, the phase encoding direction is y, and the slice encoding direction is z. The imaging matrix is 256 (x) x256 (y) x64 (z). However, the echo data actually acquired is skipped by skipping one phase encoding, and 128 phase encoding echo data are acquired. As a result, the reconstructed image is an image including a fold when the field of view (FOV) in the phase encoding direction (y) is half (that is, 15 cm). The reconstructed image is a coronal section (COR) image of the subject, and the slice direction is AP (Anterior-Posterior). Since the shooting time is TRx (phase encoding) x (slice encoding), it is ½ of normal shooting. The image data of shooting 1 is temporarily stored in, for example, an external storage device.

  Next, move the table to the top of the head 15 cm in the y direction and take a picture 2. In photographing 2, the photographing FOV 121 of the MRI apparatus is also moved to the top of the y direction by 15 cm. Therefore, as in the case of imaging 1, the imaging region coincides with the abdomen 102 of the subject. However, since the position of the RF receiving coil 405 is not changed, unlike the shooting 1, it is on the foot side of the shooting FOV 121. At this position, three-dimensional imaging of the abdomen 102 is performed with phase encoding thinned out as described above. This reconstructed image is an image including a fold when the field of view (FOV) in the phase encoding direction (y) is half (that is, 15 cm). Since this shooting time is also TRx (phase encoding) x (slice encoding), it is half that of normal shooting. The image data of shooting 2 is temporarily stored in, for example, an external storage device.

  Imaging 1 data 1 and imaging 2 data 2 have the same imaging region of the subject, both of which are folded in the phase encoding direction, and in imaging 1, the reception sensitivity distribution of the RF receiving coil 405 is unevenly distributed on the top of the head In photographing 2, the reception sensitivity distribution of the RF reception coil 405 is unevenly distributed on the foot side. Therefore, if the reception sensitivity distribution of the RF receiving coil 405 is acquired in advance, the aliasing can be removed by the aliasing removal algorithm described above. As shown in Fig. 1 (b), the reception sensitivity distribution of the abdominal radiography without folding is the reception sensitivity distribution obtained by combining the sensitivity distribution 107-1 of imaging 1 and the sensitivity distribution 107-2 of imaging 2. As in FIG. 4B, the sensitivity unevenness is reduced. On the other hand, regarding the shooting time, both the shooting 1 and shooting 2 have the phase encoding amount halved, so even if two shootings are performed, the shooting time is almost equal to the one shooting time in FIG. The table travel time is extended by about 3.5s).

  Next, move the table 15cm in sequence, and repeat the shooting, and from the shooting 3 and 4 the image without the folding of the thigh 103 (Reception sensitivity distribution that combines the sensitivity distribution 107-3 of shooting 3 and the sensitivity distribution 107-4 of shooting 4) Is obtained from the photographings 5 and 6 (having a reception sensitivity distribution obtained by combining the sensitivity distribution 107-5 of the photographing 5 and the sensitivity distribution 107-6 of the photographing 6). In shootings 3 and 5, the shooting FOV121 is returned to the -15 cm, y-direction foot side to be the same as shooting 1, and in shootings 4 and 6, the shooting FOV121 is moved 15 cm to the top in the y-direction and is the same as shooting 2. . The image data of the shootings 3 to 6 are temporarily stored in, for example, an external storage device.

The three images thus obtained are taken out of, for example, an external storage device and connected in the Y direction with a coordinate system fixed to the subject to obtain an image of the entire lower limb. As shown in the figure, the luminance distribution of the entire composite image is like a series of six peaks, which reduces the decrease in sensitivity at the connected portion. In this way, it is possible to return the sensitivity unevenness to be almost uniform, and in this method, the number of times of photographing is six times. However, since the time of each photographing is halved, photographing can be performed in the same time as before.
Note that the sensitivity range 122, subject position 123, and imaging FOV 121 in each imaging are as shown in FIG.

Next, a second embodiment of the present invention will be described. The second embodiment is a mode in which the dead time of photographing due to table movement is reduced by further moving the table in small increments for each slice encoding step in the first embodiment.
An example of the second embodiment will be described in detail with reference to FIG. The embodiment of FIG. 2 shows a case where the slice encoding is 4, as shown in FIG. However, the present invention is applicable to cases other than 4.

  The table 412 and the shooting FOV 121 are slightly shifted for each slice encoding. In other words, in 3D imaging at each station, the table 412 and the imaging FOV 121 are moved by the same interval in the crown direction for each slice encoding. In FIG. 2 (a), 3D shooting of the first station is performed in shootings 1 and 2, and the table 412 and the shooting FOV 121 are both moved by (1/8) of the shooting FOV 121 for each slice encoding. Then, in the shooting 3 of the second station, which is the next station, the shooting FOV 121 is returned to the original position (position of the slice encode 1 of shooting 1), and thereafter, the same movement as in shooting 1 and 2 is repeated.

In such shooting, the aliasing is removed in the same way as two-dimensional imaging for each slice encoding. In Fig. 2 (a), there is no aliasing in slice encoding 1 from the two measurement data of slice encoding 1 in imaging 1 and 2. A two-dimensional image in slice encoding 2 is reconstructed from the two measurement data of slice encoding 2 of photographing 1 and 2 by reconstructing a two-dimensional image. Hereinafter, similarly, two-dimensional images without wrapping in slice encodings 3 and 4 are reconstructed, respectively.

  As the reception sensitivity distribution of the RF reception coil 405 used at this time, the reception sensitivity distribution acquired by applying the same encoding amount as the acquired slice encoding amount may be used. Alternatively, the reception sensitivity distribution acquired in a state where no slice encoding is applied to all slice encoding data may be used in common in each slice encoding. In any case, the position of the shooting FOV 121 is different for each slice encoding, but the shooting FOV 121 is the same for shooting 1 and shooting 2 with the same slice encoding, so the aliasing removal algorithm applies the same idea as in 2D. it can.

  Then, after removing aliasing for each slice encoding, Fourier transform is performed in the slice encoding direction. In this Fourier transform, the influence of the positional deviation between slice encodes can be eliminated by using data at the same address in the y (phase encode) direction.

  The present embodiment has a feature that the dead time of photographing due to the movement of the table is reduced because the movement of the table is in small increments for each slice encoding. That is, the table moving distance when moving the table for each slice encoding is the shooting FOV / (2 · number of slice encodings). When the table is moved every n slice encodings, n · shooting FOV / (2 · slice encoding number). When sending a table every n slice encodings, the method of applying slice encoding in each step is 1, n + 1, 2n + 1, 3n + 1,..., Second step 2, n + 2, 2n + 2, 3n + 2, ... may be applied. Or, in the first step, 1, 2, 3, ... k: (where k = total number of slice encodings / n), in the second step k + 1, k + 2, k + 3, ... 2k, ... You may apply.

  For the sake of simplicity, FIG. 2 shows the case of shooting only twice. However, for example, as in the case of FIG. In particular, the reception sensitivity distribution in the same case as in FIG. 1 is almost as shown in the graph of FIG. 2 (b) as in FIG. 1 (b). However, in this example, since the reception sensitivity distribution in the slice direction is further different, the image uniformity is considered to be further improved.

  In the example of the MRI apparatus of FIG. 7 described above, this example is similar to the example of the first embodiment, for example, the control of the movement of the table 412 and the imaging is performed by the control unit 411, and includes the folding expansion processing. The image reconstruction process and the synthesis process of each station image are performed by the signal processing unit 407. The reception sensitivity distribution of the RF receiving coil, the measurement data acquired at each station, the storage of each station image, and the synthesized image are stored in an external storage device. Remembered.

  Note that the table movement for each slice encoding in the second embodiment can be applied to normal three-dimensional imaging without applying the parallel MRI method. In this case, normal three-dimensional imaging is performed while slightly shifting the table for each slice encoding. Then, the entire synthesized image is acquired by connecting the three-dimensional images for each station.

  In the above embodiment, the RF receiving coil 405 is a single loop coil for simplicity, but it may be a multiple array coil. As a signal synthesis method in that case, a known square sum method for improving SN can be used. In addition, if the signals of the multiple array coils are simply added by matching only the phase, pseudo QD detection can be achieved. Furthermore, parallel MRI may be applied in the slice encoding direction using multiple array coils.

  Although two embodiments of the present invention have been described above, the MRI apparatus of the present invention is not limited to the above embodiments, and various modifications can be made. For example, in the description of the above embodiment, the case where the present invention is applied to a vertical magnetic field type MRI apparatus has been described. However, the present invention is not limited to a vertical magnetic field type MRI apparatus, but may be applied to a horizontal magnetic field type MRI apparatus. Can do. However, it is necessary to adapt the shape of the RF receiving coil so that the reception sensitivity distribution matches the horizontal magnetic field method. In the above-described embodiment, the case of three stations has been described. However, the present invention is not limited to this, and the present invention can be similarly applied to two or four or more multi-station MRIs.

The figure which shows the Example of the 1st Embodiment of this invention. The figure which shows the Example of the 2nd Embodiment of this invention. The figure which shows the conventional Example (in the case of 3 times measurement) of multi-station MRI. The figure which shows the conventional Example (in the case of 5 times measurement) of multi-station MRI. The figure explaining the return of an image. The figure which shows the signal detection part of RF receiving coil to which this invention is applied. The figure which shows the whole structure of the MRI apparatus with which this invention is applied. The supplementary explanatory drawing of the signal calculation to which this invention is applied. The figure explaining the sequence of a general gradient echo.

Explanation of symbols

  401 subject, 402 magnet, 403 gradient magnetic field coil, 404 RF transmission coil, 405 RF reception coil, 406 signal detection unit, 407 signal processing unit, 408 display unit, 409 gradient magnetic field power source, 410 RF transmission unit, 411 control unit, 412 Table, 413 Head-foot (HF) direction

Claims (6)

A static magnetic field generating means for generating a uniform static magnetic field space in which the subject is arranged;
An RF receiver coil for receiving a nuclear magnetic resonance signal from the subject;
A table on which the subject is placed and moved;
Measuring means for thinning out at least phase encoding and measuring the nuclear magnetic resonance signal to obtain thinned-out k-space data;
Signal processing means for acquiring an image without aliasing using the thinned-out k-space data and the reception sensitivity distribution data of the RF receiving coil;
Control means for controlling execution of the table movement, the thinning measurement, and the image reconstruction;
Equipped with a,
The RF receiving coil is one, and is disposed in a fixed position in the uniform static magnetic field space,
The measurement means performs thinning measurement when the imaging FOV of the subject is at a first relative position with respect to the one RF receiving coil, acquires first thinning k-space data, and obtains a second relative When it is in the position, the thinning measurement is performed to obtain the second thinning k-space data,
The signal processing means includes the first thinned-out k-space data, the second thinned-out k-space data , sensitivity distribution data at the first relative position, and the second thinned-out k-space data . A magnetic resonance imaging apparatus that obtains an unfolded image related to the imaging FOV from sensitivity distribution data at a relative position of
The measurement means makes the position of the photographing FOV at the first relative position different from the position of the photographing FOV at the second relative position with respect to the uniform static magnetic field space when photographing the same photographing FOV. A magnetic resonance imaging apparatus. The magnetic resonance imaging apparatus according to claim 1.
The measurement means assigns slice encoding to measure a three-dimensional nuclear magnetic resonance signal and obtains first and second thinned-out three-dimensional k-space data at the first and second relative positions, respectively. ,
The signal processing means is configured to perform a loopback operation on the imaging FOV from the first thinned-out three-dimensional k-space data, the second thinned-out three-dimensional k-space data, and each sensitivity distribution data of the one RF receiving coil. A magnetic resonance imaging apparatus characterized by acquiring a non-existing three-dimensional image. The magnetic resonance imaging apparatus according to claim 2.
The measurement means performs the thinning measurement by moving the table at least every step of the slice encoding,
The magnetic resonance imaging apparatus, wherein the signal processing means obtains the unfolded three-dimensional image by aligning the coordinates in the moving direction of the subject and performing Fourier transform in the slice encoding direction. The magnetic resonance imaging apparatus according to claim 1.
The measuring means moves the position of the imaging FOV so that the imaging region of the subject imaged by the imaging FOV is the same between the first relative position and the second relative position. Magnetic resonance imaging apparatus. The magnetic resonance imaging apparatus according to claim 4.
The magnetic resonance imaging apparatus according to claim 1, wherein the measuring unit makes a movement amount of the position of the imaging FOV smaller than a size in a moving direction of the imaging FOV. The magnetic resonance imaging apparatus according to claim 4.
When the subject is divided into a plurality of stations and each station is imaged,
A magnetic resonance imaging apparatus, wherein the movement of the imaging FOV is repeated for each station.

Images