JP2009148635A - Magnetic resonance imaging device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an MRI device which allows parallel imaging which reduces a folding artifact and cuts down on the labor of an operator regardless of the positional relationship of a FOV (Field of View) which is arranged as an imaging domain at scanning plan and the actual domain of an object. <P>SOLUTION: An MRI device can perform parallel imaging using a multi-coil which consists of two or more component coils. The MRI device has an imaging condition setting means which sets up an imaging condition including an imaging domain of the parallel imaging (a) and specified magnification, a reconstitution means which obtains a reconstitution image (b) on the coil component to coil component basis based on a magnetic resonance signal achieved by scanning with the imaging condition, a development processing means which develops and processes the reconstitution image (b) obtained by this reconstitution means at magnification exceeding the specified magnification, and an image generation means which clips an image (d) of dimension according to the imaging domain which is set up and obtains an image for display from an image (c) obtained by this development processing means. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an MRI apparatus that performs parallel imaging (PI), which has recently attracted much attention, as one form of MR imaging utilizing a magnetic resonance (MR) phenomenon, and in particular, post-processing of parallel imaging. In particular, the present invention relates to an MRI apparatus that maximizes the deployment capability of the deployment process executed as described above and suppresses artifacts peculiar to parallel imaging.

  Magnetic resonance imaging is an imaging method in which a nuclear spin of an object placed in a static magnetic field is magnetically excited with a high-frequency signal of its Larmor frequency, and an image is reconstructed from MR signals generated by this excitation. .

  In the field of magnetic resonance imaging, researches on high-speed imaging have been particularly active in recent years. As an example of this, there is known a high-speed imaging method generally called a parallel imaging method using a multi-coil composed of a plurality of RF coils (element coils). Historically, this parallel imaging method is also called a multi-coil high-speed imaging method, a PPA (Partially Parallel Acquisition) method, or a subencoding method.

  In this parallel imaging, various modes are adopted. That is, (1) a method for calculating skipped data in k space, (2) a sub-encoding method and SENSE method for performing unfolding processing in real space, and (3) Sum of Square (SoS) which is another modified method. : Sum of squares of squares) PILS method for combining images. Among these, the methods described in Non-Patent Documents 1 to 3 and Patent Document 1 are known as methods belonging to the item (2).

  These parallel MR imaging methods basically use an array coil consisting of a plurality of RF coils (element coils) (hereinafter referred to as Phased Array Coil: PAC), a so-called multi-coil, and skip phase encoding. This is executed under the so-called sub-encoding acquisition, in which the number of phase encodings is reduced to one times the number of RF coils of a predetermined number of phase encodings necessary for image reconstruction. Thereby, each RF coil receives an echo signal simultaneously, and image data is produced | generated for every RF coil from these received echo signals. As a result, the FOV of the image generated for each RF coil is reduced, the scan time is shortened, and the shooting speed is increased.

  On the other hand, in an image obtained by reconstructing echo signals collected from each RF coil, folding (wrap-around or folding; also referred to as aliasing) occurs at the image end. Therefore, in the case of this parallel MR imaging method, using the fact that the sensitivities of the plurality of RF coils are individually different, a development process for unfolding each of a plurality of images obtained corresponding to each RF coil. Is performed as post-processing. For this development process, a spatial sensitivity map of the RF coil is used.

  A plurality of images subjected to this development processing are combined with the final full FOV (field of view) image. As described above, the parallel imaging method can speed up scanning (high-speed imaging), and can obtain a final image with a wide field of view, for example, the entire abdomen.

International Publication No. 99/054746 Pamphlet

“Ra J.B. and Rim C.Y., Fast Imaging Method Using Multiple Receiver Coils with Subencoding Data Sets, ISMRM p.1240, 10991” “Ra J.B. and Rim C.Y., Fast Imaging Using Subencoding Data Sets From Multiple Detectors, MRM 30: 142-145, 1993” “Pruessman K.P., Weiger M., Scheidegger M.B., and Boesiger P., SENSE: Sensitivity Encoding for Fast MRI, MRM 42: 952-962, 1999”

  However, in the case of the parallel imaging method described above, in principle, the images collected by the respective element coils are accompanied by the folding phenomenon as described above. Therefore, the actual region of the subject is the FOV specified in the setting of the imaging conditions. As long as it does not fit in (Field Of View), there is a problem that a folded back artifact called back folding occurs.

  In other words, in the conventional parallel imaging, the conditions for the development process are determined according to the final image (which is usually a rectangular imaging field of view) specified by the operator through the ROI, matrix, FOV conditions, etc. at the time of the scan plan, Image generation is performed under this development processing. In order to perform a satisfactory development process, it is necessary to designate ROI that strictly includes the existence area of the subject. However, in actual clinical practice, it is a great burden on the operator to always satisfy the condition. In practice, this is an inappropriate setting, causing artifacts and hindering diagnostic ability.

  This will be specifically described with reference to FIG.

  FIG. 10A shows a case where appropriate imaging can be performed even with the conventional ROI setting. The imaging condition (or scan plan) is set while the subject is completely covered with the ROI, and the sub-encoding data collection (not shown) in which the phase encoding is halved compared to the original imaging is performed. 2 times speed). Folding is an overlap of at most two points, and if a development process is performed on the assumption that the two points overlap, a desired image without the folding is obtained.

  FIG. 10B shows a case where the conventional ROI setting is not appropriate. That is, in the setting (plan) of the imaging conditions, the case where the FOV of the desired final image is specified to be smaller than the actual area of the subject is shown. In this case, when shooting is performed with the designation of the double speed, an overlap of three points occurs in part, and the folding does not disappear even by the development process premised on the double speed. FIG. 10C shows a conventional imaging example that does not use parallel imaging in the plan of FIG. In this case, since the turn-up appears at the end of the FOV, there is no problem because it does not hinder the diagnosis.

  Referring only to this figure, it is easy to perceive that it is only necessary to set a large FOV (rectangular ROI) of a desired final image in the plan. In some cases, the set ROI may protrude beyond the actual region of the subject in slices other than the slice used in step 1, and it is difficult to always set an appropriate ROI. FIG. 11 shows such an example.

  Furthermore, in the case of imaging of the heart, imaging is usually performed in accordance with the short axis, long axis, and / or the four-chamber cross section of the heart. 12A and 12B schematically show short-axis imaging of the heart. However, the above-described problems are remarkable because the cross-sectional shape of the heart itself is complicated and the cross-sections vary greatly depending on the patient. In addition, for example, since the cross-section setting is performed in accordance with the posture of the fetus, the same phenomenon occurs.

  Furthermore, when taking a coronal section of the trunk, the arm placed next to the trunk or the arms raised are wide in the left-right direction, so that it is difficult to carry out a scan plan that completely includes this.

  Thus, in many cases, planning appropriate for the development of the parallel imaging method is not always executed.

  The present invention has been made to overcome such a current situation, and the main purpose of the present invention is to return regardless of the positional relationship between the FOV set as the imaging region and the actual region of the subject at the time of the scan plan. It is possible to obtain a stable reconstructed image that is free from artifacts or reliably reduced it, and to obtain a high-quality final image, and to specify the FOV of the operator, and to reduce the effort and time required for the scan plan. It is an object of the present invention to provide an MRI apparatus for parallel imaging that can be reduced to improve patient throughput.

  In order to achieve the above object, the MRI apparatus according to the present invention has a virtual intermediate deployment (unfolding) deployment FOV having a field of view larger than the FOV set by the operator as an imaging region in the scan plan. Is automatically set on the apparatus side, and the developed FOV image is developed and created as an intermediate image. The final image corresponding to the FOV set by the operator in the scan plan is created by cutting out from the intermediate image of the developed FOV.

  Specifically, the MRI apparatus according to the present invention is based on an imaging condition setting unit that sets an imaging condition including an imaging region for parallel imaging and a specified magnification, and a magnetic resonance signal obtained by scanning under the imaging condition. Reconstructing means for obtaining a reconstructed image for each coil element, unfolding processing means for unfolding the reconstructed image obtained by the reconstructing means at a magnification exceeding the specified magnification, and obtained by the unfolding processing means. Image generating means for cutting out an image having a size corresponding to the set imaging area from the obtained image and obtaining an image for display.

  In addition, the MRI apparatus according to the present invention performs imaging FOV setting means for setting an imaging FOV (Field of View), and the parallel imaging scan using the multi-coil to perform the corresponding for each coil element. Image acquisition means for obtaining a reconstructed image of a scan, and development for obtaining the image of the developed FOV by developing the reconstructed image obtained by the image acquiring means based on a development FOV wider than the imaging FOV The image processing apparatus includes: a processing unit; and an image generation unit that obtains an image of the imaging FOV based on a developed FOV image obtained by the developing processing unit.

  According to the present invention, the “expanded expansion method”, which is a parallel imaging expansion method that virtually enlarges the area to be expanded, can be used regardless of the setting of the imaging area of the operator in the scan plan. As a result, the artifacts caused by aliasing peculiar to parallel imaging can be reliably and greatly reduced, while the burden on the operator can be reduced. The final image can be obtained stably, and the usefulness of diagnosis can be enhanced.

The functional block diagram which shows an example of the structure which concerns on embodiment of the magnetic resonance imaging apparatus of this invention. 3 is a flowchart showing an outline of parallel imaging executed in the first embodiment. The figure explaining the relationship between the change from FOV set by a scan plan to the last FOV, and the cross-sectional area | region of a subject based on the expansion expansion method of 1st Embodiment. The relationship between the change from the FOV set in the scan plan to the final FOV and the region in the case where the cross-sectional region of the subject is the short-axis cross-sectional region of the subject according to the expansion method of the first embodiment will be described. Figure. The relationship between the change from the FOV set in the scan plan to the final FOV and the region in the case where the cross-sectional region of the subject is the short-axis cross-sectional region of the subject according to the expansion method of the first embodiment will be described. Another figure. The figure explaining the relationship between the change from collection FOV to the last FOV, and a subject area | region based on the expansion expansion method combined with the magnification control based on a mask performed by the 2nd Embodiment of this invention. 6 is a flowchart showing an outline of parallel imaging executed in the second embodiment. FIG. 10 is another diagram for explaining the relationship between the change from the collection FOV to the final FOV and the subject region according to the enlargement and expansion method combined with the mask-based magnification control, which is executed according to the second embodiment of the present invention. The figure explaining the expansion expansion method combined with the bonding expansion | deployment method based on a trapezoid mask performed by the 3rd Embodiment of this invention. The figure explaining the conventional expansion | deployment process in parallel imaging. The figure explaining the inconvenience of the conventional expansion | deployment process in parallel imaging. FIG. 6 is another diagram for explaining the disadvantages of conventional development processing in parallel imaging.

  Embodiments of a magnetic resonance imaging (MRI) apparatus according to the present invention will be described below with reference to the accompanying drawings.

(First embodiment)
A first embodiment of an MR signal receiving apparatus and a magnetic resonance imaging (MRI) apparatus according to the present invention will be described with reference to FIGS. This magnetic resonance imaging apparatus will be described as a system for performing parallel imaging.

  First, an overview of the overall configuration of the magnetic resonance imaging apparatus according to this embodiment will be described with reference to FIG.

  This magnetic resonance imaging apparatus is an apparatus that can obtain an MR image by executing parallel imaging using a multi-coil. As shown in FIG. 1, this magnetic resonance imaging apparatus includes a bed unit on which a patient P as a subject is placed, a static magnetic field generation unit that generates a static magnetic field, and a gradient magnetic field generation unit for adding position information to the static magnetic field. And a transmission / reception unit that transmits and receives an RF (high frequency) signal, and a control / arithmetic unit that controls the entire system and performs image reconstruction.

The static magnetic field generation unit includes, for example, a superconducting magnet 1 and a static magnetic field power supply 2 that supplies current to the magnet 1, and an axial direction of a cylindrical opening (diagnostic space) into which the subject P is loosely inserted. The static magnetic field H ₀ is generated (corresponding to the Z-axis direction in the orthogonal coordinate axes set in the present apparatus). The magnet portion is provided with a shim coil for static magnetic field homogeneity (not shown).

  The couch portion can removably insert the top plate 14T on which the subject P is placed into the opening of the magnet 1. This insertion is performed by the bed driving device 14D. The couch driving device 14D can move the top plate 14T in the longitudinal direction (Z-axis direction) in response to a driving signal given from the host computer 6 described later. As an example, the subject P is placed along the longitudinal direction of the top 14T.

  The gradient magnetic field generating unit includes a gradient magnetic field coil unit 4G incorporated in the magnet 1. The gradient coil unit 4G includes three sets (types) of x, y, and z coils (not shown) for generating gradient magnetic fields in the X, Y, and Z axis directions orthogonal to each other. The gradient magnetic field generator further includes a gradient magnetic field amplifier 4 that supplies current to the x, y, and z coils. The gradient magnetic field amplifier 4 supplies a pulse current for generating a gradient magnetic field to each of the x, y, and z coils under the control of a sequencer 5 described later.

By controlling the pulse current supplied from the gradient magnetic field amplifier 4 to the x, y, and z coils, the gradient magnetic fields in the three orthogonal axes X, Y, and Z, which are physical axes, are combined, and the slice direction gradients are orthogonal to each other. The logical axis directions of the magnetic field Gs, the phase encoding direction gradient magnetic field Ge, and the readout direction (frequency encoding direction) gradient magnetic field Gr can be arbitrarily set and changed. Slice direction, phase encoding direction, and gradient magnetic fields in the readout direction are superimposed on the static magnetic field H _0.

  The transmission / reception unit includes a whole body (WB) coil 7T and a reception multi-coil 7R as RF coils disposed in the vicinity of the subject P in the imaging space in the bore of the magnet 1, and both the coils 7T and 7R. The transmitter 8T and the receiver 8R are connected.

  The whole body coil 7T is used as a transmission / reception coil when the whole body coil 7T is used as a single RF coil. On the other hand, when the multi-coil 7R (reception coil) is used for reception, the whole-body coil 7T is used as a transmission-only coil.

  The multi-coil 7R is configured as an array-type coil that can set a high S / N, and is formed by a plurality of RF coils 7a, 7b, 7c, and 7d, each of which forms an element coil. In the present embodiment, four (four channels) RF coils 7a, 7b, 7c, and 7d are employed, and each of the coils is formed of, for example, a circular or rectangular surface coil. As the RF coils 7a, 7b, 7c, and 7d, coils having an appropriate size are used so that a desired FOV (region of interest) can be covered with the four-channel coils.

  The outputs of the four-channel RF coils 7a, 7b, 7c, and 7 are sent to the receiver 8R independently of each other.

  1 schematically illustrates that the four-channel RF coils 7a to 7d of the multi-coil 7R shown in FIG. 1 are arranged along the body surface of the subject P. However, the multicoil 7R is not necessarily limited to a structure composed of a plurality of surface coils, and may be composed of a plurality of volume coils or a QD coil. Further, the multi-coil may be attached to the bed or may be attached so as to be attached to the subject.

  The transmitter 8T and the receiver 8R operate under the control of the sequencer 5 described later. The transmitter 8T supplies a Larmor frequency RF current pulse for causing nuclear magnetic resonance (NMR) to the magnetization spin of the subject P to the whole body coil 7T. The receiver 8R takes in an echo signal (high-frequency signal) received by the whole-body coil 7T or the multi-coil 7R and generates echo data (original data).

  Specifically, as shown in FIG. 1, the receiver 8R is divided into a receiving unit on the whole body coil side and a receiving unit on the multi-coil side.

  The reception unit on the whole body coil side includes a duplexer 81 connected to the whole body coil 7T, a preamplifier 82 connected to the duplexer 81, and a reception system circuit 83 that receives a reception signal of the preamplifier 82. A transmitter 8T is also connected to the duplexer 81.

  As a result, the duplexer 81 passes the transmission drive pulse from the transmitter 8T toward the whole body coil 7T at the time of transmission, and passes the echo signal detected by the whole body coil 7T toward the preamplifier 82 at the time of reception. The preamplifier 82 preamplifies the received echo signal and sends it to the reception system circuit 83. The reception system circuit 83 performs various signal processing such as intermediate frequency conversion, phase detection, low frequency amplification, and filtering on the input echo signal, and then performs A / D conversion to generate echo data (original data). This is sent to the host computer 6.

  On the other hand, in the receiving section on the multi-RF coil side, the four-channel signals transmitted from the four-channel RF coils 7a to 7d are respectively transmitted to the reception system circuits 86A to 86D. Similarly to the above, each of the receiving system circuits 86A to 86D performs various signal processing such as intermediate frequency conversion, phase detection, low frequency amplification and filtering on the input echo signal, and then performs A / D conversion. Generate echo data. The signals thus subjected to reception processing are sent from the reception system circuits 86A to 86D to the host computer 6.

  Further, the control / arithmetic unit includes a sequencer (also called a sequence controller) 5, a host computer 6, a storage device 11, a display device 12, and an input device 13.

  Among them, the host computer 6 uses the sequencer 5 for various preparation scans, scan plans, and imaging books based on the software procedure stored in the internal memory or the storage device 11. Perform scan (imaging scan) and post-processing. Among these, the prescan for preparation includes a pilot scan for positioning, a shimming scan for homogenizing the static magnetic field, a sensitivity map scan for measuring a sensitivity map of each element coil used for the unfolding process, and the like. At the time of this preparation pre-scan and main scan, the host computer gives the pulse sequence information necessary for these scans to the sequencer 5.

  Further, the host computer 6 functions as an interface with the operator together with the display device and the input device 13 during the scan plan, and the operator can interactively input the scan plan information to the apparatus side.

  In particular, the host computer 6 has a function of estimating sensitivity maps of the RF coils 7a to 7d of the multi-coil 7R, a function of performing reconstruction processing on echo data and calculating image data, and a function of controlling the driving of the bed driving device 14D. Etc. The imaging conditions include information on the phase encoding direction of parallel MR imaging and the position, size, and shape of the FOV.

  The scan based on the pulse sequence information is a scan for collecting a set of echo data necessary for image reconstruction. As the pulse sequence, a three-dimensional (3D) scan or a two-dimensional (2D) scan) sequence is used. The pulse train forms include SE (spin echo) method, FSE (fast SE) method, FASE (fast Asymmetric SE) method (that is, imaging method combining the fast SE method with the half Fourier method), EPI (echo planar imaging). ) Method, FE (gradient echo) method, FFE (fast FE) method, segmented FFE method, etc. are used.

  The sequencer 5 includes a CPU and a memory, stores pulse sequence information sent from the host computer 6, and controls operations of the gradient magnetic field amplifier 4, the transmitter 8T, and the receiver 8R according to this information. The pulse sequence information is all information necessary to operate the gradient magnetic field amplifier 4, the transmitter 8T, and the receiver 8R according to a series of pulse sequences. For example, the pulse sequence information includes the pulse current applied to the x, y, and z coils. Contains information on strength, application time, application timing, and the like.

  The data of the image reconstructed in the parallel imaging and the synthesized image data are displayed on the display device 12 and stored in the storage device 11. Information relating to imaging conditions desired by the operator, pulse sequence, image composition and difference calculation is input to the host computer 6 via the input unit 13.

  Next, the FOV setting principle related to the development processing method executed in the parallel imaging of this embodiment will be described.

  This “development processing method” is different from the conventional development processing method, and has a larger area independently of the “set FOV” (= “FOV of the final image”) set in the scan plan. This is a method of automatically setting “development FOV” as “intermediate FOV”, performing image data processing in the middle of this development FOV, and finally cutting out the image of “final FOV” from the image of “development FOV”.

  The relationship between these FOVs is as follows.

“Setting FOV” = L_final: FOV indicating a desired imaging area designated by the operator in the imaging condition setting (scan plan). At this time, the acceleration rate R (parallel imaging magnification) is also designated.
“Acquisition FOV” = L_acquis: FOV for each element coil to collect an echo signal, which is determined by L_acquis = L_final / R.
“Development FOV” (intermediate FOV) = L_unfold: L_unfold> L_final = R · L_acquis is set.
“Final FOV”: For example, an FOV corresponding to “setting FOV”. An image of this final FOV is cut out from an image of “development FOV”.

  When this is compared with the conventional expansion processing method, in the conventional case, “set FOV” = “expanded FOV” = “final FOV” is set.

  The present invention is not limited to such a premise, and by setting “deployment FOV”> “final FOV” (= “setting FOV”, for example), as described below, a large merit is obtained in practical use, that is, as described above. It tries to overcome the conventional problems.

  Note that a technique for virtually enlarging a region to be expanded according to the present invention is referred to as an “expanded unfolding technique”.

The expansion ratio R ′ (= L_unfold / L_aquis) based on this expansion expansion method is therefore:
[Equation 1]
R '= L_unfold / L_aquis> L_final / L_aquis = (R ・ L_acquis) / L_aquis = R
It becomes.

  That is, the expansion ratio R ′ is set so as to satisfy R ′> R. For example, the expansion ratio R ′ is an integer satisfying R ′> R. For example, Int (R) +1 (Int: integer part) may be used. For example, when R = 2, R ′ = 3, 4,. When R = 1.5, R ′ = 2, 3,. Alternatively, R ′ may be the upper limit of R ′ = Nc (the number of reception channels). However, if R ′ is increased too much, the g factor increases, leading to a decrease in SNR, and the sensitivity map of each element coil is required for the expansion process, so the value of R ′ is simple. In addition, it is necessary to make it smaller than the existence range of the sensitivity map for performing expansion in the entire region. The g factor is an index representing the SNR decrease due to the expansion process (see the document “Pruessman K, et al., SENSE: Sensitivity Encoding for Fast MRI, MRM 42: 952-962, 1999”).

  An example of the overall operation of parallel imaging employing the “enlargement and expansion method” will be described with reference to FIGS.

  During this parallel imaging, the host computer 6 executes processing in the order shown in FIG. The host computer 6 first performs patient registration based on input information from the operator (step S1), and then executes a pilot scan for positioning using the whole body coil 7T to obtain a positioning image (step S2). In addition, when this positioning image has already been collected, the positioning image can also be used. Next, the host computer 6 executes a shimming scan (step S3), and further executes a sensitivity map scan for collecting sensitivity map data of the element coils 7a to 7d (step S4). Note that this sensitivity map scan does not necessarily have to be performed prior to the main scan for imaging described later, and may be performed as part of a series of pulse sequences of the main scan.

  When the prescan for preparation is completed in this way, the host computer 6 makes an interactive scan plan (setting of imaging conditions) with an operator who inputs necessary information while observing the positioning image (step S5A). .

  Specifically, in addition to normal parameters such as the type of pulse sequence, a speed-up rate (parallel imaging magnification) R and a FOV (setting FOV) set via an ROI indicating a desired imaging region are set. Accept (step S5A). In response to this, the host computer 6 automatically calculates the collected FOV as described above and stores the information (step S5B). Next, the host computer 6 reads an appropriate expansion magnification R ′ set in advance, automatically calculates the expansion FOV using the expansion magnification R ′ as described above, and stores the information (steps S5C, S5D). ).

  When this is finished, the host computer 6 causes the sequencer 5 to execute parallel imaging in a desired mode as the main scan under the collection FOV (step S6), and the echo signal collected from each element coil 7a (˜7d) An image in real space is obtained by reconstructing the echo data based on it (step S7).

  Each reconstructed image corresponding to each element coil 7a (up to 7d) is developed, and a desired final FOV image is cut out from the developed image (steps S8 and S9).

  This unfolding process and cutting process will be described with reference to FIG. This figure shows the case where the speed-up rate (magnification) R = 2. For this reason, parallel imaging with a speed-up rate R = 2 × speed with respect to a set FOV indicating a desired imaging area set on the positioning image (indicating a state set smaller than the actual area of the subject) (Sub-encoding data collection) is performed, and each element coil 7a (˜7d) collects data from the collection FOV for ½ matrix (see FIGS. 3A and 3B).

  On the other hand, in the present embodiment, the expansion magnification R ′ (> R) is automatically set on the apparatus side (the operator does not need to be aware of this), and for example, the expansion magnification R ′ = 3 × speed. An unfolding process (unfolding process) is performed on the FOV using the sensitivity map data of each element coil 7a (˜7d) (see FIG. 3C). In the case of this unfolding FOV with the unfolding magnification R ′ = 3 × speed, the general imaging surely includes the actual region of the subject. Therefore, the host computer 6 cuts out an image having a desired set FOV size from the real space image of the developed FOV for each pixel, and creates a final FOV image (see FIG. 3D). This cutting process is also automatically performed on the apparatus side (the operator does not need to be aware of this).

  The real space image of the final FOV created in this way is displayed on the display device 12 and the image data is stored in the storage device 11 (step S10). As a result, the operator can visually observe the FOV image set in the scan plan, and does not need to be aware of the processing associated with the previous expansion method.

  FIGS. 4 and 5 show examples in which the expansion processing method illustrated in FIG. 3 is applied to a cross section in the short axis direction of the heart. In either case, the speed-up rate R = 2 times and the development magnification R ′ = 3 times, and the heart short-axis cross section protrudes beyond the FOV set at the time of the first scan plan. In particular, in the case of FIG. 5, the set FOV is too small compared to that of FIG. 4, and the extent of such protrusion is larger. However, the desired final FOV image is obtained in any case by employing the enlargement and expansion method.

  As described above, in the present embodiment, by using the enlargement and expansion method in which the region used for the expansion process is set virtually large, an image in which the occurrence of the aliasing residual artifact called backfolding is reliably eliminated or reduced is generated once. The final FOV image can be obtained by cutting out from this image. For this reason, even when the imaging area (setting FOV) set in the scan plan is incomplete due to the relationship with the existing area of the subject, that is, even if the existing area protrudes from the imaging area, the development FOV is set. Development processing under optimum conditions can be performed, and high-quality images with reduced artifacts can be provided.

  In other words, even if the imaging area (setting FOV) set in the scan plan is rough to some extent, the enlargement / expansion method is executed automatically by internal processing (particularly without notifying the operator) on the apparatus side. For this reason, the operator does not need to be as nervous as in the conventional method for specifying the imaging region, and labor and time for the specification can be saved, and this burden is remarkably reduced.

(Second Embodiment)
Next, a second embodiment according to the present invention will be described with reference to FIGS. The second embodiment is characterized in that the above-described enlargement / development method is performed in combination with magnification control based on mask information. The hardware configuration of the magnetic resonance imaging apparatus is the same as that of the first embodiment.

  First, magnification control based on mask information will be described. In the first embodiment, the expansion and expansion method has been described, but the number of aliasing that substantially overlaps in the collection matrix is different at each point.

  Hereinafter, a point set that is assumed to be a substantial real area of the subject will be referred to as a “mask” for the sake of simplicity. The description will be made assuming that the shape of the subject in FIG. 6 is a mask as it is.

  In FIG. 6, at the point A in the “collection FOV”, the corresponding three points A1, A2, and A3 in the “deployment FOV” are folded back, and it is necessary to actually perform the triple speed development. As for the point B, the three points B1, B2, and B3 overlap, but the point B1 shown in parentheses can be removed from the calculation because there is no signal, and the expansion can be substantially doubled. Therefore, the number of points (equivalent points) where the aliasing overlaps in the estimated existing region of the subject or a predetermined extended region including the same is calculated in advance with reference to the mask information described above. In the enlargement / expansion method, enlargement / expansion is performed for each pixel at a magnification according to the number of equivalence points. In other words, if the equivalence score of a certain pixel is 2, double expansion is performed, and even if the equivalence score of another pixel is 3, even if it is on the same image, double expansion is performed. Constitute. “Final FOV” creates an image of a portion cut out according to the setting FOV specified in the scan plan.

  In order to use this mask processing together, the host computer 6 executes parallel imaging in the general procedure shown in FIG. Compared with the procedure shown in FIG. 2 described above, this procedure is characterized by the scan plan (imaging plan) in step S5 and the expansion process in step S8, and other steps are the same as or equivalent to those in FIG. It has become.

  In other words, in the scan plan, the host computer 6 sets the imaging conditions, sets the collection FOV, reads the development magnification, and sets the development FOV (steps S5A to S5D), and then uses the echo data collected in the pilot scan as the subject. The mask data for estimating the real area is created (step S5E). Next, the host computer 6 calculates the number of points where the aliasing overlaps (has the same value) (the number of equivalent values) for each pixel from the mask data and the acceleration rate R (step S5F). Then, in the expansion process executed after the main scan and the image reconstruction (steps S6 and S7), the above-described expansion / development process with reference to the equivalence score is executed as described above.

  As a result, the point A in FIG. 6 in which the folding of substantially three points has occurred (corresponding to the point that cannot be properly developed by the conventional method) is expanded three times in the expansion method according to the present embodiment. Is done. Further, since the point B is substantially a turn of two points, the enlargement and expansion method is limited to double expansion, and image quality deterioration (SNR decrease) due to excessive expansion is prevented.

  As described above, by controlling the expansion ratio for each pixel based on the mask information, the three-fold expansion gives priority to preventing the aliasing, while the two-fold expansion works to prevent the SNR from decreasing due to the increase of the g factor. Yes. Therefore, the expansion will work very effectively.

  The enlargement / expansion method combined with the magnification control using the mask exhibits a more remarkable effect in the case shown in FIG. That is, although the subject protrudes from both ends of the set FOV, the number of substantial overlapping points is at most two at any point. Here, representative two points C and D are shown. Not only this point, but double-speed expansion processing is sufficient at any point, so that the SNR reduction is suppressed within the optimum range, and a final image without aliasing can be obtained.

(Third embodiment)
Next, a third embodiment according to the present invention will be described with reference to FIG. The third embodiment relates to a process for avoiding image quality discontinuity in the above-described enlargement and expansion method combined with magnification control based on mask information. The hardware configuration of the magnetic resonance imaging apparatus is the same as that of the first embodiment.

  In the second embodiment described above, the method of changing the double speed rate for each point in the expansion and expansion method has been shown. When the speed factor is changed, the g factor changes discontinuously, so that the image information may change discontinuously for each pixel. Therefore, in order to avoid discontinuity in image quality due to discontinuous change in the expansion rate (integer) of the development, a method of continuously weighting and adding two images with different double rates at the boundary (here, This is referred to as “Glued Unfolding Technique”.

  In the present embodiment, a method of performing this bonding development method using a trapezoidal mask (“trapezoidal mask method”) is shown. As shown in FIG. 9, a mask that covers the external shape of the subject is M, and a trapezoidal mask having a value of 1 on an appropriate region M ′ including M and having a trapezoidal function shape in the phase encoding direction. “Think T.

  Therefore, the host computer 6 additionally calculates the following processing in “collection FOV” sequentially in, for example, steps S5B and S8 of FIG. 7 described above, and calculates the result value E (yk) as “expanded FOV”. The above value.

  [1] Corresponding point coordinates yk, k = 1, 2,. Let R1 be the number of T values = 1, R2 be the number of 0 <T values <= 1, and r = ΣT (yk) (0 ≦ R1 ≦ r ≦ R2).

[2] For yk with T (yk) = 0, the obtained value E (yk) = 0. Assuming that yk satisfying T (yk)> 0, E (yk) is obtained for those points.
(1) When 0 <R1 = R2 (that is, a set of points of t (yk) = 1):
R1 is expanded, and the expanded value is defined as E1 (yk).
The value E (yk) to be obtained for the corresponding yk: = E1 (yk).
(2) When 0 = R1 <R2 (that is, a set of points of T (yk) <1):
R2 is expanded and the expanded value is E2 (yk).
The value E (yk) to be obtained for the corresponding yk: = T (yk) * E2 (yk).
(3) In the case of 0 <R1 <R2 (that is, a set in which T (yk) = 1 and T (yk) <1 are mixed): R1 expansion and R2 expansion are performed, and the expanded value is E1 Let (yk) and E2 (yk).
For the corresponding yk, the required value E (yk) is determined as follows:
[Equation 2]
When T (yk) = 1
E (yk): = ((r−R1) / (R2−R1)) * E1 (yk)
+ ((R2-r) / (R2-R1)) * E2 (yk)
When T (yk) <1,
E (yk): = T (yk) * E2 (yk)
This method shows a method of changing the weight linearly, but the weight may be changed in a sine wave shape.

  The trapezoidal mask method is a practical method capable of including all the contents of the subsequent continuous bonding process in the function value.

  The present invention is not limited to the description of the embodiment and the modifications thereof, and those skilled in the art can appropriately change and modify the invention without departing from the scope of the claims.

DESCRIPTION OF SYMBOLS 1 Magnet 3 Gradient magnetic field coil unit 5 Sequencer 6 Host computer 7T Coil for whole body 7R for transmission as RF coil Multi coil 7a-7d for reception as RF coil Surface coil 8T as element coil Transmitter 8R Receiver 13 Input 85A-85D Preamplifier 86A-86D Reception system circuit

Claims (5)

In an MRI (magnetic resonance imaging) apparatus capable of performing parallel imaging using a multi-coil composed of a plurality of element coils,
An imaging condition setting means for setting an imaging condition including an imaging area of the parallel imaging and a specified magnification;
Reconstructing means for obtaining a reconstructed image for each coil element based on a magnetic resonance signal obtained by scanning under the imaging conditions;
Development processing means for developing the reconstructed image obtained by the reconstruction means at a magnification exceeding the specified magnification;
An MRI apparatus comprising: an image generation unit that obtains an image for display by cutting out an image having a size corresponding to the set imaging region from an image obtained by the expansion processing unit. In an MRI (magnetic resonance imaging) apparatus capable of performing parallel imaging using a multi-coil composed of a plurality of element coils,
An imaging FOV setting means for setting an imaging FOV (Field of View);
Image acquisition means for performing a scan of the parallel imaging using the multi-coil to obtain a reconstructed image of the scan for each coil element;
Development processing means for performing a development process on the reconstructed image obtained by the image acquisition means based on a development FOV wider than the imaging FOV, and obtaining an image of the development FOV;
An MRI apparatus comprising: an image generation unit that obtains an image of the imaging FOV based on an image of the development FOV obtained by the development processing unit. The MRI apparatus according to claim 2,
A collection FOV determination means for determining a collection FOV when performing the parallel imaging scan on the subject based on the imaging FOV;
The MRI apparatus, wherein the image acquisition unit is configured to perform the parallel imaging scan using the multi-coil based on the acquired FOV determined by the acquired FOV determining unit. The MRI apparatus according to claim 2,
The MRI apparatus, wherein the image generation means is configured to cut out the image of the developed FOV with the imaging FOV. The MRI apparatus according to claim 2,
A real area estimation means for estimating the real area of the subject in real space;
The expansion processing means obtains a folding equivalence score in the reconstructed image obtained by the image acquisition means for each pixel in the estimated real area of the subject or an extended area including the real area, and the folding equivalence score An MRI apparatus configured to execute the expansion processing for each pixel based on the above.

Images