JP2005334101A - Magnetic resonance imaging equipment - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide MRI equipment capable of providing a dynamic EPI (Echo Planar Imaging) image with a favorable image quality by eliminating image distortion and its fluctuation, image displacement and its fluctuation, and artifact caused by the vibration of a cryocooler when performing the dynamic EPI. <P>SOLUTION: This MRI equipment previously measures three-dimensional fluctuation data indicating a period of a static magnetic field fluctuation, the amount of fluctuation, and positional dependence of the amount of fluctuation and stores them in a storage means in a computer corresponding to the vibration phase of the cryocooler. This equipment corrects the displacement for every pixel of the image using the three-dimensional fluctuation data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a magnetic resonance imaging (hereinafter abbreviated as “MRI”) apparatus for taking a tomographic image of a desired part of a subject using a nuclear magnetic resonance phenomenon, and in particular, a static magnetic field fluctuation caused by a cryocooler. The present invention relates to a technique for correcting image quality degradation during EPI dynamic shooting.

The superconducting MRI apparatus includes a superconducting magnet as a static magnetic field generation source. This superconducting magnet is an electromagnet having a superconducting coil cooled to a cryogenic temperature. The superconducting coil is disposed in a helium vessel filled with liquid helium and kept at a cryogenic state. A cryocooler is directly connected to the helium container to cool the evaporated helium gas and return it to liquid helium again.
The cryocooler usually has components that move in a piston cycle of 1 to 1.5 seconds. The mechanical vibration is transmitted to the superconducting magnet, and the vibration of the superconducting magnet fluctuates the static magnetic field. A horizontal magnetic field (that is, cylindrical) MRI apparatus has the same amount of fluctuation everywhere in the imaging space. However, in a vertical magnetic field type open MRI apparatus, the magnitude of the static magnetic field fluctuation varies depending on the location in the imaging space. is there.

  Further, a high-performance cryocooler uses a magnetic material as a piston-moving component, and the piston motion itself of the magnetic material may cause a static magnetic field fluctuation. Also in this case, the magnetic field near the cryocooler fluctuates greatly, and the magnetic field far from the cryocooler becomes small fluctuation, which may cause a difference in fluctuation amount depending on the place in the imaging space.

  When dynamic imaging using the EPI (Echo Planer Imaging) method is performed with an MRI apparatus with static magnetic field fluctuations as described above, image position displacement between EPI images, ie, image Distortion occurs. In particular, in multi-shot EPI, image blur occurs in addition to image position shift. This is because when the divided shots are subjected to different static magnetic field fluctuations, the divided number of images are formed at different positions in the phase encoding direction, and the resulting images are combined, i.e., blurred. Such an image is generated. In addition, a positional shift for each location also occurs in each image.

Furthermore, if the amount of change in the static magnetic field differs in the shooting space, a phenomenon occurs in which the image distortion differs between EPI images.
In particular, dynamic imaging using the EPI method is used for Functional MRI (hereinafter abbreviated as `` fMRI ''), which extracts the local activity of the brain from slight signal changes, but this fMRI is a slight difference in local signal intensity. Therefore, if a positional deviation or distortion occurs in the image, an inaccurate measurement result is obtained.

As a technique for correcting the positional shift due to the static magnetic field fluctuation, for example, there is a technique described in [Patent Document 1]. This corrects a positional shift caused by a static magnetic field variation using signal data having a phase encode amount of zero among measurement data acquired by EPI imaging. However, this technique is not limited to static magnetic field fluctuations, and it is possible to correct image misalignment caused by characteristic fluctuations of the MRI apparatus that occur as the operating time of the MRI apparatus elapses.
Further, as another technique for correcting the positional shift due to the static magnetic field fluctuation, there is a method of stopping a cryocooler that is a vibration generation source during photographing.
JP-A-9-289980

In the technique described in [Patent Document 1], the positional deviation correction is performed using the phase value of the signal with the phase encoding amount zero in the measurement data acquired by the EPI method, so that it is constant in the imaging space. This is intended to correct misalignment caused by static magnetic field fluctuations in the imaging space, and is not considered for misalignment correction caused by static magnetic field fluctuations that vary locally in the imaging space.
Furthermore, the technique described in [Patent Document 1] is a technique applied to the one-shot EPI, and the application to multi-shot has not been studied in detail.

  In addition, as described above, the cryocooler is normally continuously operated for 24 hours regardless of the use or non-use of the MRI apparatus in order to suppress the evaporation amount of the refrigerant (liquid helium) that cools the superconducting coil inside the superconducting magnet. It should be. Therefore, it is not economically preferable to stop the cryocooler during photographing because the amount of evaporation of the expensive refrigerant increases during that time.

  Furthermore, the amount of fluctuation of the static magnetic field fluctuation caused by the cryocooler may vary depending on the position in the imaging space. The technique described in [Patent Document 1] uses a phase value of a signal having a phase encoding amount of zero among measurement data acquired by the EPI method as described above, and therefore, a constant static value in the imaging space. Although it is possible to correct misalignment due to magnetic field fluctuations, no consideration is given to static magnetic field fluctuations having different fluctuation amounts in places within the imaging space.

  Accordingly, the present invention has been made to solve the above-described problems, and the object of the present invention is to correct misalignment in response to static magnetic field fluctuations that are different in places in an imaging space caused by a cryocooler. Is to do.

In order to achieve the above object, the MRI apparatus of the present invention is configured as follows. That is,
A static magnetic field generating means for generating a static magnetic field in the imaging space; a measurement control means for measuring an echo signal corresponding to the k-space from a subject arranged in the imaging space based on an EPI sequence; and an image from the echo signal. A signal processing means for reconfiguring, wherein the static magnetic field generating means includes a superconducting coil and a helium container filled with liquid helium for maintaining the superconducting coil at a cryogenic temperature. In a magnetic resonance imaging apparatus provided with a cryocooler that condenses vaporized helium gas and returns it to liquid helium again, means for detecting the vibration phase of the cryocooler, and three-dimensional fluctuation of the static magnetic field based on the vibration of the cryocooler Storage means for storing three-dimensional correction data for correcting the amount in association with the vibration phase of the cryocooler; Wherein the vibration phase of the cooler in response to the position of the imaging section of the specimen, to correct the positional deviation generated in the image using the three-dimensional correction data.

  As a result, the image misalignment on the image caused by the cryocooler can be corrected in accordance with the vibration phase of the cryocooler and the position of the imaging plane, thereby improving the image quality. Can do. In particular, in dynamic EPI photography, it is easily affected by the cryocooler vibration, so that it is possible to obtain high-quality images without depending on the vibration while the cryocooler is always operated, and the liquid helium indirectly. Consumption can be reduced.

In addition, a preferred embodiment of the MRI apparatus of the present invention is a multi-shot type in which the EPI sequence is divided into a plurality of shots so as to obtain the echo signal corresponding to each region of the divided k space. And
The signal processing means performs the positional deviation correction for each of the images reconstructed from the echo signals acquired in the respective shots, and then combines the corrected images to form one image. get.

  As a result, even in the multi-shot EPI sequence, as in the case of the above effect, the positional deviation on the image caused by the cryocooler corresponds to the vibration phase of the cryocooler and the position of the imaging plane. Misalignment can be corrected and image quality can be improved.

  According to the present invention, in a superconducting MRI apparatus, it is possible to obtain a dynamic imaged EPI image with good image quality by correcting misalignment, image distortion, and image blur on an image due to a static magnetic field variation caused by a cryocooler. Be able to.

  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 embodiments of the invention, and the repetitive description thereof is omitted.

  First, an overall outline of an MRI apparatus to which the present invention is applied will be described with reference to FIG. As shown in FIG. 1, the present invention will be described by taking an MRI apparatus using a vertical magnetic field type superconducting magnet as an example, but the present invention is also applicable to an MRI apparatus using a horizontal magnetic field type superconducting magnet. . Compared with a horizontal magnetic field type superconducting magnet, a vertical magnetic field type superconducting magnet is more likely to have a large amount of vibration of the static magnetic field variation caused by the cryocooler and the amount of vibration depending on the position. This is more suitable for an MRI apparatus using a superconducting magnet of the type.

  The subject 102 is placed in a magnet 101 that generates a uniform static magnetic field, and a high-frequency magnetic field necessary for generating nuclear magnetic resonance is generated in the transmission / reception system 106 for hydrogen nuclei (protons) in the subject 102. The high-frequency magnetic field is applied to the subject 102 placed on the top plate 117 from the upper transmission coil 103 and the lower transmission coil 104. After irradiation for a certain period of time, a nuclear magnetic resonance signal generated by hydrogen nuclei in the subject 102 is detected by the reception coil 105, and the detected nuclear magnetic resonance signal is converted to an audible frequency by the transmission / reception system 106, and further A / D It becomes a digital signal by the converter 107. The digital signal is processed by the operation control unit 109 to be reconstructed into an image. In addition, the gradient magnetic field for adding position information necessary for imaging includes a gradient magnetic field power supply 108 controlled by the operation control unit 109 so as to satisfy predetermined necessary conditions, and an upper gradient connected to the power supply. It is generated by the magnetic field coil 114 and the lower gradient magnetic field coil 115. The transmission / reception system 106 is similarly controlled by a computer. This computer is provided with storage means such as a hard disk or a magneto-optical disk, and stores correction data to be described later.

  The magnet 101 has a helium container 110 and a heat shielding plate 111 disposed therein, and a superconducting coil 112 is disposed in the helium container 110. The helium vessel 110 is filled with liquid helium, which is a refrigerant for cooling the superconducting coil, and the superconducting coil 112 is kept at a very low temperature. The superconducting coil 112 has a resistance of almost zero in an extremely low temperature state, and once the current flows, the current can be maintained for a long time. The heat shielding plate 111 is arranged to prevent radiant heat from being transferred to the helium vessel 110 from the outside. The cryocooler 113 is installed directly connected to the helium vessel 110 in order to cool the heat shielding plate 111 and cool the evaporated helium gas and to liquefy it again into liquid helium.

  The cryocooler usually has a component that moves in a piston cycle of 1 to 1.5 seconds, and the mechanical vibration is transmitted to the magnet, and the vibration of the magnet causes the static magnetic field fluctuation. A high-performance cryocooler uses a magnetic body as a piston-moving component, and the piston movement of the magnetic body itself causes fluctuations in the static magnetic field.

  In particular, since a pair of upper and lower helium containers are connected to each other in the vertical magnetic field type MRI apparatus, the cryocooler 113 is installed in the upper helium container 110. Therefore, the vibration of the cryocooler 113 is directly transmitted to the upper helium container 110 to cause the vibration of the helium container. This vibration of the helium container 110 fluctuates the static magnetic field in the imaging space. Moreover, the amount of fluctuation in the static magnetic field closer to the cryocooler 113 is larger, and the amount of fluctuation in the static magnetic field gradually decreases as the distance from the cryocooler 113 increases. This is because the lower helium container is fixed to the floor surface, and therefore there is less vibration than the upper helium container.

  FIG. 2 shows a specific example of the static magnetic field fluctuation. The static magnetic field fluctuation amounts at positions 201 to 203 in the imaging space are shown in graphs 211 to 213, respectively. At the position 201 close to the cryocooler, the static magnetic field fluctuates greatly. For example, as shown at 211, it fluctuates periodically with an amplitude of 40 nT (nano tesla). At the center 202 of the imaging space, the amount of change in the static magnetic field decreases, and periodically changes with an amplitude of 15 nT, for example, as indicated by 212. Further, the static magnetic field fluctuation amount is small at a substantially lower position 203 far from the cryocooler in the imaging space, and periodically fluctuates with an amplitude of 1.5 nT, for example, as indicated by 213. Since the vibration source is common, the period and phase of the static magnetic field fluctuation coincide with each other in the imaging space.

  When one-shot EPI imaging is performed with an MRI apparatus having such a static magnetic field fluctuation, an image position shift occurs in the phase encoding direction according to the amount of the static magnetic field fluctuation. Therefore, in the dynamic shooting by the one-shot EPI under such a situation, a positional deviation occurs between the EPI images. In particular, in fMRI, since a slight difference in local signal intensity is detected, an image having such a positional deviation is inappropriate for fMRI.

  Furthermore, the problem becomes complicated if the amount of change in the static magnetic field varies from place to place in the imaging space. Since the amount of positional deviation in the phase encoding direction differs depending on the location, an image having a different shape distortion at each location is generated as a result. The image captured by EPI has image distortion due to the non-uniformity of the static magnetic field, but if dynamic shooting is performed with EPI while there is static magnetic field fluctuation, the image distortion will also change. It becomes an image unsuitable for fMRI.

  In the case of multi-shot EPI shooting, the problem is further complicated. If each divided shot is subjected to different static magnetic field fluctuations, the divided number of images will be imaged at different positions in the phase encoding direction, and they will be combined, that is, blurred An image is generated. In addition, a positional shift for each location, that is, image distortion also occurs in each image.

  Next, the present invention will be described. In the present invention, the period of the static magnetic field fluctuation caused by the vibration of the cryocooler and the position dependency of the vibration amount, that is, the static magnetic field fluctuation amount for each location in the imaging space are measured in advance as three-dimensional fluctuation data. Thus, for example, it is stored in a storage means (for example, a hard disk or an optical disk) in the computer, and image positional deviation correction is performed based on the three-dimensional fluctuation data.

(First embodiment)
First, a first embodiment of the present invention will be described. In the present embodiment, the positional deviation correction of the present invention is applied to an image obtained by one-shot EPI imaging. An example of this embodiment will be described in detail below based on the processing flow shown in FIG.
Before performing one-shot EPI photography, the correction data of the period, fluctuation amount, and position dependence of the fluctuation amount of the static magnetic field fluctuation caused by the cryocooler vibration is measured in advance and stored in the computer. Store that information.

  The correction data such as the position dependence of the static magnetic field fluctuation amount, for example, is associated with the cryocooler vibration phase at intervals of about 50 mm using a small phantom, and for each location in the imaging space for each cryocooler vibration phase. It is obtained by measuring the amount of static magnetic field fluctuation. As described above, since the cryocooler vibrates mechanically with a period of 1 to 1.5 seconds, the static magnetic field also varies with a period of 1 to 1.5 seconds. Therefore, for example, one cycle is divided into 10 and the static magnetic field fluctuation amount (that is, the three-dimensional static magnetic field fluctuation amount) is associated with the vibration phase of the cryocooler at that time for each location in the imaging space at intervals of 0.1 to 0.15 seconds. Measure and use three-dimensional fluctuation data. Note that the three-dimensional fluctuation data at the phase other than the measurement phase can be obtained by interpolation from the measurement phase data without actually measuring.

Here, the vibration phase of the cryocooler 113 can be detected by, for example, attaching an acceleration sensor to the cryocooler 113 and detecting the vibration. Since the vibration of the cryocooler 113 is constant, it is not necessary to detect vibration throughout the shooting. If the vibration phase at a certain time before the shooting is detected and the subsequent shooting is performed based on the time, any arbitrary shooting can be performed. The vibration phase of the cryocooler 113 at this time can be calculated.
In addition, the signed relative value of the static magnetic field fluctuation is stored in the in-computer storage means as three-dimensional fluctuation data.

In step 301, the vibration phase of the cryocooler 113 is acquired. This is actually measured or calculated from the actual measurement value as described above.
In step 302, dynamic shooting is performed with one-shot EPI, and an echo signal of a specified shooting plane is acquired. As described above, the vibration phase of the cryocooler 113 during photographing can be calculated by detecting the vibration of the cryocooler 113 before photographing.

  In step 303, the position correction is performed by applying the position correction technique described in [Patent Document 1]. The correction method described in this document is based on the phase value of the echo center in the echo signal with the phase encode amount zero, and the time difference (TE) between the application time of the high frequency magnetic field and the echo center in the echo signal with the phase encode amount zero. The phase rotation amount per unit time in the echo signal is estimated, the phase correction amount of each echo signal having a non-zero phase encoding amount is determined from the phase rotation amount per unit time, and the phase correction of each echo signal is performed. Is. By this phase correction, it is possible to correct the displacement of the entire image.

This process will be described more specifically. The phase Phs0 at the center of the echo is obtained from an echo signal having a phase encoding amount of zero, and the position shift correction amount of the entire image is obtained from this. In other words, the positional deviation correction amount L of the entire image is ω≡Phs0 / TE (1)
Δθ≡ (ω × Δt) (2)
L≡ (FOV / M) × (ω × Δt) / 2π (3)
As required. Here, ω represents the time change rate (so-called angular velocity) of the phase received by the echo signal in the one-shot EPI of step 302, which is the phase Phs0 of the echo center of the echo signal whose phase encoding amount is zero, and the application of the high-frequency magnetic field It is a value divided by the time interval TE (that is, echo time) from the time to that. FOV is a field of view. Δt represents a time interval between adjacent echo signals in the one-shot EPI, and Δθ represents a phase rotation amount within the time interval, that is, a phase change amount in the echo signal for each phase encoding. M is the number of matrices in the phase encoding direction of the image.

Generally, it is known that the first-order term of the phase change in the phase encoding direction in k space (that is, the amount of phase change in the echo signal for each phase encoding) corresponds to the positional deviation of the image in the phase encoding direction in the image space. . Therefore, the positional deviation correction of the entire image is performed by giving each echo signal a phase rotation corresponding to the phase encoding amount. That is, for an echo signal with a phase encoding amount j (for example, an integer value of −128 to 127),
θ (j) ≡− (Phs0 + Δθ × j) (4)
Gives the phase rotation. Due to the phase rotation for each echo signal, the entire image can be shifted by -L in the phase encoding direction to correct the overall positional deviation.

  In step 304, fluctuation data corresponding to the cryocooler vibration phase acquired in step 301 and the position of the imaging plane set during imaging in step 302 are acquired from the database 310 in which the three-dimensional fluctuation data is stored.

In step 305, the fluctuation data measured in advance in the imaging plane acquired in step 304 is normalized in the plane to obtain fluctuation relative value data, and a positional deviation correction value for each pixel of the image is obtained therefrom.
In order to obtain the fluctuation relative value data of the imaging plane, the sum of the fluctuation data for each pixel where the imaging target exists in this plane is obtained, and the value obtained by dividing the fluctuation data for each pixel by this total is obtained. Normalized correction value. Specifically, assuming that the variation data for each pixel is Vij (i, j is pixel coordinates), the normalized correction value for each pixel is expressed as Vij / (ΣVij / N) (5)
It can be. Here, N is the total number of calculation target pixels. Accordingly, the actual misregistration correction amount Pij for each pixel is calculated by using the misregistration correction amount L of the entire image in Expression (3). Pij≡L × {Vij / (ΣVij / N)} (6)
It can be expressed as.

  The normalization process will be specifically described with reference to FIG. In FIG. 4, imaging plane variation data 402 as shown in FIG. 4 (b) is extracted from the three-dimensional variation data 401 measured and stored in advance as shown in FIG. 4 (a). The photographing plane variation data 402 is obtained by two-dimensionally arraying variation data for each pixel. Each numerical value represents the amount of positional deviation in the phase encoding direction for each pixel. Therefore, if the subject to be imaged is present in the four pixel regions surrounded by the thick square 403, the sum of the regions is 44, and the average is 11. Therefore, a value obtained by dividing each pixel value by the average value 11 is a normalized correction value for each pixel. This is shown in FIG. 4 (c).

In step 306, from the misregistration correction amount L of the entire image obtained in step 303 and the misregistration correction value Pij for each pixel obtained in step 305, in addition to the misalignment correction of the entire image, the necessary phase encoding direction A misregistration correction value Cij is obtained for each pixel. This value can be calculated from equations (3) and (6)
Cij≡L-Pij (7)
It becomes.

In equation (7), since the entire image positional deviation correction has already been performed in step 303, in order to perform the positional deviation correction for each pixel obtained in step 305, how much positional deviation correction should be performed for each pixel. Represents. If Cij is a positive value, the pixel is shifted in the positive phase encoding direction. Conversely, if Cij is negative, the pixel is shifted in the negative phase encoding direction.
In step 307, based on the correction value Cij for each pixel in the imaging plane obtained in step 306, the positional deviation correction for each pixel is performed on the imaging plane image.

  The positional deviation correction for each pixel in steps 306 and 307 will be specifically described using the example shown in FIG. FIG. 5 specifically shows the positional deviation of the four pixel regions 403 where the object to be photographed exists and the correction for each pixel in the photographing plane shown in FIG.

When the cryocooler vibration state is 501 and the echo signal 502 at the maximum value is detected and an image of the imaging plane is obtained based on the echo signal, the echo at the minimum value is 505. A case where the signal 503 is detected and an image of the imaging plane is obtained based on the echo signal is shown at 504.
Reference numerals 504-1 and 505-1 denote pixel regions 403 with no positional deviation that should be obtained when the cryocooler does not vibrate.

  Reference numerals 504-2 and 505-2 denote cases in which image distortion due to positional deviation for each pixel occurs in the pixel region 403 due to the vibration of the cryocooler. In 504-2, four pixels were displaced in the phase encoding direction by (+ L) × 30/11, (+ L) × 10/11, (+ L) × 1/11, and (+ L) × 3/11, respectively. In the case of 505-2, the four pixels are (−L) × 30/11, (−L) × 10/11, (−L) × 1/11, (−L) × 3/11, respectively. Only the case where a positional deviation occurs in the phase encoding direction is shown.

  504-3 and 505-3 are the cases where the position correction of the entire image performed in step 303 is performed, and 504-3 is shifted by + L in the positive phase encoding direction as a whole. In order to correct it as a whole, the positional deviation correction is performed by −L in the negative phase encoding direction, and the 505-3 is shifted by −L in the negative phase encoding direction as a whole. This is a case where the positional deviation is corrected by + L in the positive phase encoding direction in order to correct the error as a whole. In other words, in 504-3, each of the four pixels is shifted by -L in the negative phase encoding direction, and in 505-3, each of the four pixels is shifted by + L in the positive phase encoding direction. Is the case.

  Reference numerals 504-4 and 505-4 denote cases where positional deviation correction is performed for each pixel. In 504-4, in order to correct the positional deviation for each pixel indicated by 504-2, only the correction that cannot be corrected by the positional deviation correction of the entire image performed in 504-3 is performed. That is, four pixels are phase-encoded by (−L) × 19/11, (+ L) × 1/11, (+ L) × 10/11, and (+ L) × 8/11, respectively, based on equation (7) This is a case where the positional deviation correction is performed after shifting to. Similarly, in 505-4, in order to correct the positional deviation for each pixel indicated by 505-2, only the amount that cannot be corrected by the positional deviation correction of the entire image performed in 505-3 is performed. That is, the four pixels are shifted by (+ L) × 19/11, (−L) × 1/11, (−L) × 10/11, (−L) × 8/11, respectively, and positioned in the phase encoding direction. This is a case where deviation correction is performed.

  The pixel shift process will be specifically described with reference to FIG. The values of the lattice points 601 to 605 in the phase encoding direction of the image obtained in the state where the positional deviation has occurred are the values of the positions 611 to 615 that are not originally the lattice points because the positional deviation has occurred. It should be. Therefore, the values of the obtained grid points 601 to 605 are shifted by the shift amounts obtained in Steps 303 and 307, respectively, and the values of points 611 to 615 are obtained as the interpolation curves passing through these shifted points 611 to 615. Ask for. For example, a known spline interpolation method or the like can be used. Then, the values of the grid points 601 to 605 are obtained from the obtained interpolation curve. In this way, the value on the lattice point subjected to the positional deviation correction is obtained, and the positional deviation correction in the phase encoding direction of the captured planar image can be performed for each pixel.

(Second Embodiment)
Next, a second embodiment of the present invention will be described. In this embodiment, the positional deviation correction according to the present invention is applied to an image obtained by multi-shot EPI imaging. An example of this embodiment will be described in detail with reference to the processing flow shown in FIG.
As in the first embodiment, before performing multi-shot EPI imaging, the period of the static magnetic field fluctuation, the amount of fluctuation, and the correction data for the position dependence of the quantity of fluctuation are measured in advance. Then, the information is stored in the storage means in the computer. However, since the cryocooler vibration data can be the same as that at the time of a single shot, there is no need to acquire it again if it has already been acquired.

  In the case of multi-shot EPI imaging, the cryocooler vibration phase varies from shot to shot, so the amount of static magnetic field variation varies from shot to shot, and as a result, the amount of positional deviation in the phase encoding direction also varies. Therefore, when the positional deviation correction method of the present invention is applied to multi-shot EPI imaging, it is necessary to obtain the vibration phase of the cryocooler for each shot and perform positional deviation correction corresponding to the vibration phase. However, as with single shots, the cryocooler's vibrations are constant, so there is no need to detect vibrations during shooting, and the vibration phase at a certain time before shooting is detected, and subsequent shootings are based on that time. If done, the phase of the cryocooler at any time during imaging can be obtained by calculation.

For this purpose, an image is reconstructed from only the echo signal for each shot, and image misalignment correction is performed on the image in the same manner as in the first embodiment. After combining, the final image is obtained.
Each processing step of misalignment correction in this multi-shot EPI imaging will be described in detail with reference to FIG.

In step 701, an initial value 1 is set to a loop counter n for performing dynamic shooting with multi-shot EPI.
In step 702, the vibration phase of the cryocooler is acquired as in step 301. Or calculate by calculation.
In step 703, an echo signal in the nth shot is measured.

In step 704, the cryocooler vibration phase acquired in step 702 and the n-th shot echo signal measured in step 703 are associated with each other and temporarily stored in the in-computer storage means.
In step 705, it is determined whether or not n has reached the maximum number of shots. If not, the process proceeds to step 706, and if it has reached, the process proceeds to step 707.

In step 706, the loop counter is incremented and the process proceeds to step 702.
In step 707, since all the k-space data necessary for image reconstruction has been prepared, the same positional deviation correction of the entire image as in step 303 is performed. However, Phs0 and Δθ in the equation (4) use values acquired in shots including a phase encoding amount of zero in other shots.
In step 708, a loop counter for performing positional deviation correction for each pixel of the image for each shot is set to an initial value 1.
In step 709, the image is reconstructed only from the echo signal corresponding to the nth shot.

In step 710, pixel-by-pixel misalignment correction is performed on the image reconstructed from only the echo signal corresponding to the nth shot in step 709. For this purpose, the following processes are performed. That is,
As in step 304, the cryocooler vibration phase at the time of echo signal measurement in this shot acquired in step 702 and the fluctuation data corresponding to the position of this imaging plane are stored in the database 310 in which the three-dimensional fluctuation data is stored. Get from
Similar to step 305, the fluctuation data is normalized in the imaging plane to obtain fluctuation relative value data, and from this, the positional deviation correction value for each pixel of the image is obtained,
Similar to step 306, in addition to the amount of misalignment correction for the entire image, further necessary misalignment correction values in the phase encoding direction are obtained for each pixel and temporarily stored in the in-computer storage means.
As in step 307, the positional deviation correction in the phase encoding direction is performed for each pixel. However, in the process after n = 2, the misalignment correction value for each pixel obtained when n = 1 is used.

  In step 711, the image corrected for positional deviation for each pixel in step 710 is subjected to inverse Fourier transform to obtain k-space data, which is temporarily stored in the in-computer storage means.

In step 712, it is determined whether or not n has reached the maximum number of shots. If not, the process proceeds to step 713, and if it has reached, the process proceeds to step 714.
In step 713, the loop counter is incremented and the process proceeds to step 709.
In step 714, the temporarily stored k-space data is synthesized (for example, added on the k-space), and Fourier transformed to reconstruct an image.

In the present invention, since the three-dimensional fluctuation data of the static magnetic field fluctuation amount is measured in advance and stored in the computer storage means, the phase of the echo center of the echo signal whose phase encoding is zero and the time from the high-frequency magnetic field until then From the interval (TE), it is also possible to predict the amount of static magnetic field fluctuation when acquiring an echo signal with a non-zero phase encoding.
That is, if the static magnetic field fluctuation amount is obtained for a shot in which an echo signal with zero phase encoding exists, the static magnetic field fluctuation amount in a shot without an echo signal with zero phase encoding can also be estimated. In the above processing flow, the cryocooler vibration phase is acquired for each shot in the above processing step 702, but this is performed only for the shot including the measurement of the echo signal whose phase encoding is zero. The vibration phase of the cryocooler may be estimated from the time interval between the shot and the other shots, and this estimated phase may be used.

The figure which shows the outline of MR apparatus of this invention. The figure which shows the mode of the static magnetic field fluctuation | variation in imaging | photography space. The figure which shows the processing flow in the single shot EPI of this invention. (a) is a figure which shows three-dimensional fluctuation | variation data, (b) is a figure which shows the specific example of normalization of fluctuation | variation data within an imaging plane. The figure which shows the outline of the positional offset correction for every pixel. The figure which shows the outline of the process which shifts pixel data to a phase encoding direction at the time of performing position shift correction for every pixel. The process by the multi-shot EPI of this invention is shown.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 101 ... Magnet, 102 ... Subject, 103 ... Upper transmission coil, 104 ... Lower transmission coil, 105 ... Reception coil, 106 ... Transmission / reception system, 107 ... A / D converter, 108 ... Gradient magnetic field power supply, 109 ... Computer, 110 ... Helium vessel, 111 ... Heat shield, 112 ... Superconducting coil, 113 ... Cryocooler, 114 ... Gradient magnetic field coil, 115 ... Lower gradient magnetic field coil, 116 ... Subject space, 117 ... Top plate

Claims (2)

A static magnetic field generating means for generating a static magnetic field in the imaging space; a measurement control means for measuring an echo signal corresponding to k-space from a subject arranged in the imaging space based on an EPI sequence; and an image from the echo signal. Comprising signal processing means for reconfiguration,
The static magnetic field generating means includes a superconducting coil and a helium container filled with liquid helium for maintaining the superconducting coil at a cryogenic temperature.
The helium container is a magnetic resonance imaging apparatus including a cryocooler that condenses helium gas vaporized from liquid helium and returns it to liquid helium again.
Means for detecting a vibration phase of the cryocooler;
Storage means for storing three-dimensional correction data for correcting the three-dimensional fluctuation amount of the static magnetic field based on the vibration of the cryocooler in association with the vibration phase of the cryocooler;
A magnetic resonance imaging apparatus that corrects a positional shift generated in the image using the three-dimensional correction data in correspondence with a vibration phase of the cryocooler and a position of an imaging section of the subject. The magnetic resonance imaging apparatus according to claim 1.
The EPI sequence is a multi-shot type divided into a plurality of shots so as to obtain the echo signal corresponding to each of the divided areas of the k space,
The signal processing means performs the positional deviation correction for each of the images reconstructed from the echo signals acquired in the respective shots, and then combines the corrected images to form one image. A magnetic resonance imaging apparatus characterized by comprising:

Images