JP2013017811A - Magnetic resonance imaging apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To calculate actual gradient magnetic field waveforms in MRI with a high precision, and to perform regridding processing with a high precision on the basis of the gradient magnetic field waveform.SOLUTION: An MRI apparatus includes a gradient magnetic field calculating unit and a regridding processing unit. The gradient magnetic field calculating unit calculates a waveform of a gradient magnetic field current supplied to a gradient magnetic field coil on the basis of conditions of an imaging sequence and calculates the gradient magnetic field waveform in a readout direction on the basis of a mutual inductance by which the gradient magnetic field coil causes the mutual induction and the waveform of the gradient magnetic field current. The regridding processing unit generates or rearranges k-space data so that a part of a nuclear magnetic resonance signal acquired during a time zone in which a time integral value of an intensity of the gradient magnetic field in the readout direction is non-linear may undergo sampling, and the time integral value up to a sampling period corresponding to each matrix element in the k-space data may equally be spaced.

Description

  Embodiments of the invention relate to magnetic resonance imaging.

  MRI is an imaging method in which a nuclear spin of a subject placed in a static magnetic field is magnetically excited with an RF pulse having a Larmor frequency, and an image is reconstructed from MR signals generated by the excitation. The MRI means magnetic resonance imaging, the RF pulse means radio frequency pulse, and the MR signal means nuclear magnetic resonance signal. .

  In MRI, in order to obtain spatial position information, gradient magnetic fields orthogonal to each other are applied superimposed on a static magnetic field. Therefore, the gradient magnetic field generation system of the MRI apparatus includes a gradient magnetic field coil that adds spatial position information to the MR signal by applying a gradient magnetic field in the imaging space where the subject is placed.

  For example, when reconstructing a two-dimensional image, three gradient magnetic fields are used: a slice selection direction gradient magnetic field, a phase encoding direction gradient magnetic field, and a readout direction gradient magnetic field. The gradient magnetic field usually has a pulse-like waveform and is also called a gradient magnetic field pulse. The waveform and amplitude of the gradient magnetic field pulse are defined as a part of imaging sequence parameters determined by imaging conditions and the like.

  Among the gradient magnetic field pulses, the readout direction gradient magnetic field pulse applies a magnetic field having a gradient determined by the amplitude of the gradient magnetic field pulse in the readout direction. The MR signal (echo signal) emitted from the subject is sampled during the application of the read direction gradient magnetic field pulse, that is, during the ON period of the pulse. If the amplitude of the gradient magnetic field pulse during the ON period is constant, the gradient of the magnetic field in the readout direction is constant, and a linear relationship between the position in the readout direction and the frequency of the MR signal is ensured.

  In the high-speed imaging method, sampling in the readout direction is performed in a short period. For example, in a high-speed imaging method called EPI (Echo Planer Imaging), scanning is performed at high speed and continuously while reversing the direction of the gradient magnetic field in the readout direction for one magnetic excitation (MR signal acquisition). Is done.

  The pulse waveform of the EPI readout direction gradient magnetic field has a shorter pulse width and shorter pulse period than other imaging methods. That is, the frequency component of the pulse waveform of the gradient magnetic field in the EPI readout direction is higher than that of other imaging methods.

  On the other hand, the gradient magnetic field pulse is generated by applying a pulsed current to the gradient coil. Although the shape of the current pulse applied to the gradient coil is ideally a rectangular wave, it is actually a trapezoidal wave having a rising region and a falling region. As a result, the pulse waveform of the gradient magnetic field is not an ideal rectangular wave but a trapezoidal wave having a rising region and a falling region.

  In general, in a high-speed imaging method such as EPI, the pulse width of a gradient magnetic field pulse is short, and the ratio of the rising and falling regions at both ends of the pulse to the entire pulse width is high. For this reason, a method has been proposed in which data is sampled not only in a flat pulse region but also in the rising and falling regions and used as data for image reconstruction.

  A method of sampling data in the rising and falling regions is called Ramp Sampling. Ramp Sampling shortens the data collection period compared to sampling only a region where the gradient magnetic field strength is flat.

  However, the raw data of the MR signal sampled at equal intervals in the rising region and the falling region is sampled when the gradient magnetic field is changed, so that it is not evenly spaced in the k space. Therefore, it is desirable to rearrange the sampled raw data so as to be equally spaced in the k space before reconstruction. This reordering process is usually called regridding.

  In the regridding process described in Patent Document 1, it is assumed that the waveform of the gradient magnetic field pulse is not a simple trapezoidal waveform but a nonlinear waveform, and the waveform of the nonlinear gradient magnetic field pulse is calculated based on the waveform of the gradient magnetic field current. Is done. In the method of Patent Document 1, the regridding process is executed based on the waveform of the gradient magnetic field pulse.

JP 2010-172383 A

  The prior art of Patent Document 1 is based on the premise that the waveform of a gradient magnetic field pulse is similar to the waveform of a current (hereinafter referred to as a gradient magnetic field current) supplied to a gradient magnetic field coil. That is, when the gradient magnetic field waveform is non-linear, the gradient magnetic field pulse waveform is similar to the non-linear current waveform.

  Then, the gradient magnetic field current is actually measured by an ammeter, and the regridding process or the like is executed based on the gradient magnetic field pulse having a shape similar to the measured current waveform. In addition, there is also a technology for calculating an output current waveform based on an input signal (control signal) to the gradient magnetic field power supply by simulation or the like, and performing a gridding process based on a gradient magnetic field pulse having a shape similar to the calculated output current waveform. It is disclosed.

  However, the waveform of the gradient magnetic field current does not necessarily match the waveform of the gradient magnetic field actually generated from this gradient magnetic field current.

  For this reason, in MRI, a technique for calculating an actual gradient magnetic field waveform with high accuracy and executing the regridding processing or parameter correction processing with high accuracy based on the calculated gradient magnetic field waveform has been desired.

(1) In one embodiment, the MRI apparatus generates k-space data composed of a plurality of matrix elements by applying a gradient magnetic field to the imaging region and sampling MR signals collected from the imaging region. The image data is reconstructed based on the k-space data. The MRI apparatus includes a gradient magnetic field power source, a gradient magnetic field calculation unit, and a regridding processing unit.
The gradient magnetic field power source applies a gradient magnetic field to the imaging region by causing a gradient magnetic field current to flow through the gradient coil according to the imaging sequence.
The gradient magnetic field calculation unit calculates the waveform of the gradient magnetic field current based on the conditions of the imaging sequence, and the gradient magnetic field in the readout direction based on the mutual inductance in which the gradient magnetic field coil causes mutual induction and the waveform of the gradient magnetic field current. Calculate the waveform.
The regridding processing unit samples a portion of the MR signal collected in a time zone in which the time integral value of the gradient magnetic field strength in the reading direction is nonlinear, and a sampling period corresponding to each matrix element The k-space data is generated or rearranged so that the time integration values up to are equally spaced.

(2) In another embodiment, the MRI apparatus applies a gradient magnetic field to the imaging region and reconstructs image data based on MR signals collected from the imaging region. This MRI apparatus includes a gradient magnetic field power source similar to (1) above, a gradient magnetic field calculation unit similar to (1) above, and a waveform correction unit.
The waveform correction unit captures a part of the conditions of the imaging sequence so that the gradient magnetic field waveform approaches the target waveform when the gradient magnetic field waveform calculated by the gradient magnetic field calculation unit is different from the target waveform of the gradient magnetic field waveform. Correct before executing the sequence.

1 is a block diagram showing the overall configuration of an MRI apparatus in a first embodiment. The functional block diagram of the computer 58 of FIG. The circuit diagram which shows an example of the equivalent circuit model of a gradient magnetic field generation system. The graph which represented typically the measured value of the frequency characteristic of real part Re {Z} of the impedance Z of a gradient magnetic field coil. The graph which represented typically the measured value of the frequency characteristic of Im {Z} / (omega) which divided the imaginary part Im {Z} of the impedance Z of the gradient magnetic field coil by the angular frequency (omega). The schematic diagram which shows the concept of the conventional regridding process. The schematic diagram which shows the concept of the calculation method of the gradient magnetic field in 1st Embodiment. The schematic diagram which shows an example of the imaging sequence of EPI of a spin echo system. The schematic diagram which shows an example of the data of MR signal before converting into k space data, when the number of matrixes of phase encoding and frequency encoding is 256x256. The conceptual diagram which shows that MR signal sampled at equal intervals in time becomes a non-equal interval on k space in the area | region where the reading direction gradient magnetic field is nonlinear. The schematic diagram which shows the concept of the 1st method of the regridding process in 1st Embodiment. The schematic diagram which shows the concept of the 2nd method of the regridding process in 1st Embodiment. 6 is a flowchart showing an example of an operation flow of the MRI apparatus of the first embodiment. The flowchart which shows an example of the flow of operation | movement of the MRI apparatus of 2nd Embodiment. The schematic diagram which shows an example of the correction method of the parameter regarding the slice selection direction gradient magnetic field in EPI. The schematic diagram which shows an example of the correction method of the parameter regarding the phase encoding direction gradient magnetic field in EPI. The flowchart which shows an example of the flow of operation | movement of the MRI apparatus of 3rd Embodiment. The flowchart which shows an example of the flow of operation | movement of the MRI apparatus of 4th Embodiment. The circuit diagram which shows another example of the equivalent circuit model of a gradient magnetic field generation system. The circuit diagram which shows another example of the equivalent circuit model of a gradient magnetic field generation system.

  Hereinafter, embodiments of an MRI apparatus and an MRI method will be described with reference to the accompanying drawings. In addition, in each figure, the same code | symbol is attached | subjected to the same element and the overlapping description is abbreviate | omitted.

(First embodiment)
FIG. 1 is a block diagram showing the overall configuration of the MRI apparatus 20 in the first embodiment. The hardware configuration of the MRI apparatus 20 shown in FIG. 1 and FIG. 2 described later is common to the first to fourth embodiments.

  As shown in FIG. 1, the MRI apparatus 20 includes a cylindrical static magnetic field magnet 22 for forming a static magnetic field, a cylindrical shim coil 24 provided with the same axis inside the static magnetic field magnet 22, A gradient magnetic field coil 26, an RF coil 28, a control device 30, and a bed 32 on which a subject QQ is placed are provided.

  Here, as an example, the X-axis, Y-axis, and Z-axis that are orthogonal to each other in the apparatus coordinate system are defined as follows. First, the static magnetic field magnet 22 and the shim coil 24 are arranged so that their axial directions are orthogonal to the vertical direction, and the axial direction of the static magnetic field magnet 22 and the shim coil 24 is the Z-axis direction. Further, it is assumed that the vertical direction is the Y-axis direction, and the bed 32 is arranged such that the normal direction of the surface for placing the top plate is the Y-axis direction.

  The control device 30 includes, for example, a static magnetic field power supply 40, a shim coil power supply 42, a gradient magnetic field power supply 44, an RF transmitter 46, an RF receiver 48, a sequence controller 56, and a computer 58.

  The gradient magnetic field power supply 44 includes an X-axis gradient magnetic field power supply 44x, a Y-axis gradient magnetic field power supply 44y, and a Z-axis gradient magnetic field power supply 44z. The computer 58 includes an arithmetic device 60, an input device 62, a display device 64, and a storage device 66.

  The static magnetic field magnet 22 is connected to the static magnetic field power supply 40 and forms a static magnetic field in the imaging space by the current supplied from the static magnetic field power supply 40.

  The imaging space means, for example, a space in the gantry where the subject QQ is placed and a static magnetic field is applied. The gantry is a structure formed, for example, in a cylindrical shape so as to include the static magnetic field magnet 22, the shim coil 24, the gradient magnetic field coil 26, and the RF coil 28. 1 is complicated, the components such as the static magnetic field magnet 22 in the gantry are illustrated, and the gantry itself is not illustrated.

  The shim coil 24 is connected to a shim coil power source 42 and makes the static magnetic field uniform by a current supplied from the shim coil power source 42.

  The static magnetic field magnet 22 is often composed of a superconducting coil, and is connected to the static magnetic field power source 40 and supplied with current when excited, but after being excited, it is disconnected. Is common. The static magnetic field magnet 22 may be formed of a permanent magnet without providing the static magnetic field power supply 40.

  The gradient magnetic field coil 26 includes an X-axis gradient magnetic field coil 26 x, a Y-axis gradient magnetic field coil 26 y, and a Z-axis gradient magnetic field coil 26 z, and is formed in a cylindrical shape inside the static magnetic field magnet 22. The X-axis gradient magnetic field coil 26x, the Y-axis gradient magnetic field coil 26y, and the Z-axis gradient magnetic field coil 26z are connected to the X-axis gradient magnetic field power source 44x, the Y-axis gradient magnetic field power source 44y, and the Z-axis gradient magnetic field power source 44z, respectively.

  By means of currents respectively supplied from the X axis gradient magnetic field power supply 44x, the Y axis gradient magnetic field power supply 44y, and the Z axis gradient magnetic field power supply 44z to the X axis gradient magnetic field coil 26x, the Y axis gradient magnetic field coil 26y, and the Z axis gradient magnetic field coil 26z, A gradient magnetic field Gx in the X-axis direction, a gradient magnetic field Gy in the Y-axis direction, and a gradient magnetic field Gz in the Z-axis direction of the coordinate system are formed in the imaging region, respectively.

  That is, the gradient magnetic fields Gx, Gy, and Gz in the three-axis directions of the apparatus coordinate system are synthesized, and the slice selection direction gradient magnetic field Gss, the phase encode direction gradient magnetic field Gpe, and the readout direction (frequency encode direction) as the logical axes. Each direction of the magnetic field Gro can be arbitrarily set. The gradient magnetic fields Gss, Gpe, and Gro in the slice selection direction, the phase encoding direction, and the readout direction are superimposed on the static magnetic field.

  The “imaging region” means, for example, a region that is an image and is at least part of an MR signal collection range used for generating one image or one set of images. The imaging area is defined in terms of position and range as a part of the imaging space, for example, by the apparatus coordinate system. When the entire MR signal collection range is an image, that is, the MR signal collection range and the imaging region may completely match, but both may not match completely. For example, when MR signals are collected in a wider range than the region to be an image in order to prevent aliasing artifacts, the imaging region can be said to be a part of the MR signal collection range.

  The “one image” and “one set of images” may be two-dimensional images or three-dimensional images. “One set of images” refers to “multiple images” when MR signals of a plurality of images are collected in one pulse sequence, such as multi-slice imaging.

  The RF transmitter 46 generates an RF pulse (RF current pulse) with a Larmor frequency for causing nuclear magnetic resonance based on the control information input from the sequence controller 56 and transmits this to the RF coil 28 for transmission. To do. The RF coil 28 includes a whole-body coil for transmitting and receiving RF pulses built in the gantry, and a local coil for receiving RF pulses provided in the vicinity of the bed 32 or the subject QQ.

  The transmission RF coil 28 receives an RF pulse from the RF transmitter 46 and transmits it to the subject QQ. The receiving RF coil 28 receives an MR signal (high frequency signal) generated by exciting the nuclear spin inside the subject QQ by the RF pulse, and this MR signal is detected by the RF receiver 48. .

  The RF receiver 48 performs various signal processing such as pre-amplification, intermediate frequency conversion, phase detection, low-frequency amplification, and filtering on the detected MR signal, and then performs A / D (analog to digital) conversion. The raw data that is the digitized complex data is generated. The RF receiver 48 inputs the generated raw data of the MR signal to the sequence controller 56.

  The arithmetic device 60 performs system control of the entire MRI apparatus 20, and will be described with reference to FIG.

  The sequence controller 56 stores control information necessary for driving the gradient magnetic field power supply 44, the RF transmitter 46, and the RF receiver 48 in accordance with a command from the arithmetic device 60.

  The control information here includes, for example, the intensity and application time and application timing of the gradient magnetic field pulse current to be applied to the gradient magnetic field power supply 44, and the intensity and application time and application timing of the RF pulse output from the RF transmitter 46. Is sequence information describing operation control information related to a pulse sequence such as.

  The sequence controller 56 drives the gradient magnetic field power supply 44, the RF transmitter 46, and the RF receiver 48 according to the stored predetermined sequence, thereby causing the X-axis gradient magnetic field Gx, the Y-axis gradient magnetic field Gy, the Z-axis gradient magnetic field Gz, and the RF. Generate a pulse. Further, the sequence controller 56 receives the raw data of the MR signal input from the RF receiver 48 and inputs this to the arithmetic device 60.

  The computing device 60 of the computer 58 realizes various functions in accordance with programs stored in the storage device 66 in addition to the control of the sequence controller 56 and the system control of the entire MRI apparatus 20. These various functions may be realized by providing a specific circuit in the MRI apparatus 20 without depending on a program.

  The configuration (gantry) other than the control device 30 is usually installed in an examination room, and the configuration of the control device 30 is often installed in a room (for example, a machine room) different from the examination room. However, the embodiment of the present invention is not limited to such an arrangement. For example, the RF receiver 48 may be arranged in the gantry. An RF receiver 48 is disposed in the gantry in the vicinity of the receiving RF coil 28, and an analog signal is converted into a digital signal (and further converted into an optical signal) in the RF receiver 48, and the machine room sequence is converted. The data may be transmitted to the controller 56. In this case, unnecessary noise can be reduced.

FIG. 2 is a functional block diagram of the computer 58 shown in FIG.
As shown in FIG. 2, the arithmetic device 60 of the computer 58 includes a sequence controller control unit 300, an imaging condition setting unit 302, a gradient magnetic field calculation unit 306 (including a circuit constant storage unit 308), a constant correction unit 312, and a program. It functions as a waveform correction unit 316, an image reconstruction unit 90 (including a k-space database 92 and a gridding processing unit 320), an image database 94, an image processing unit 96, a display control unit 98, and the like.

  The imaging condition setting unit 302 sets imaging sequence conditions including parameter values such as a pulse width and a pulse amplitude based on instruction information from the input device 62 and a determination result regarding the calculated gradient magnetic field waveform.

  The imaging condition setting unit 302 sets an imaging sequence based on various sequence conditions input via the input device 62 and inputs the imaging sequence to the sequence controller control unit 300.

  The sequence controller control unit 300 inputs the imaging sequence set by the imaging condition setting unit 302 to the sequence controller 56.

  Magnet for static magnetic field 22, static magnetic field power supply 40, shim coil 24, shim coil power supply 42, gradient magnetic field coil 26, gradient magnetic field power supply 44, RF transmitter 46, RF receiver 48, sequence controller 56, sequence controller control unit 300, imaging condition A data collection unit is configured by the setting unit 302 and the like.

  A pre-scan described later is executed by the data collection unit. Further, the data collection unit executes the main scan according to the imaging sequence set by the imaging condition setting unit 302, and thereby MR signals from the subject QQ are collected as raw data.

  The image reconstruction unit 90 converts the raw data of the MR signal input from the RF receiver 48 via the sequence controller 56 into k-space data and stores it in the k-space (frequency space) in the k-space database 92.

Further, the image reconstruction unit 90 takes in k-space data from the k-space database 92 and performs image reconstruction processing including Fourier transform on the k-space data to generate image data.
The image reconstruction unit 90 stores the generated image data in the image database 94.

  The image reconstruction unit 90 includes a regridding processing unit 320 that performs a regridding process.

  The image processing unit 96 reads necessary image data from the image database 94, and performs display processing such as image processing such as difference processing and MIP processing (maximum intensity projection processing) on the image database 94 for display. Generate image data. The image processing unit 96 stores the generated image data in the storage device 66.

  The display control unit 98 performs control for displaying the display image data stored in the storage device 66 and characters and images for various user interfaces on the display device 64.

  Similar to the first embodiment, the gradient magnetic field calculation unit 306 performs the following calculation under the assumption that there is a coil (coil that is electromagnetically coupled to the gradient magnetic field coil 26) different from the gradient magnetic field coil 26. Execute the process.

That is, the gradient magnetic field calculation unit 306 is based on the current flowing through the gradient magnetic field coil 26 in an equivalent circuit (for example, see FIG. 3 described later) of the gradient magnetic field generation system including the gradient magnetic field coil 26 and the “other coil”. Then, the current (I ₁ (t) and I ₂ (t) in FIG. 3) flowing in the “other coil” is calculated. In the following description, the current (Iout (t)) flowing through the gradient magnetic field coil 26 is referred to as “gradient magnetic field current”.

The gradient magnetic field calculation unit 306 is configured to generate gradient magnetic field waveforms (for example, the X axis, Y axis, and Z axis gradients) formed in the imaging region based on the gradient magnetic field current and the current flowing in the “other coil”. (Time change of magnetic fields Gx, Gy, Gz) is further calculated.
The gradient magnetic field calculation unit 306 includes a circuit constant storage unit 308 that stores circuit constants in the equivalent circuit.

  The constant correction unit 312 acquires an evaluation image by a prescan described later. When there is an artifact in the evaluation image, the constant correction unit 312 corrects the circuit constant of the equivalent circuit so that the artifact is reduced.

  Since the first embodiment will be described with a focus on the regridding process, correction of circuit constants will not be described. For this reason, a specific function (correction of circuit constants) of the constant correction unit 312 will be described in the second embodiment below.

  When the gradient magnetic field waveform calculated by the gradient magnetic field calculation unit 306 is different from the setting value of the gradient magnetic field waveform, the waveform correction unit 316 corrects the imaging sequence parameter to bring the gradient magnetic field waveform closer to the target waveform. .

  Since the first embodiment will be described focusing on the gridding process, the correction of imaging sequence parameters will not be described. Therefore, a specific function of the waveform correction unit 316 (processing for bringing the gradient magnetic field waveform close to the target waveform) will be described in a third embodiment to be described later.

  The input device 62 is, for example, an input device such as a keyboard and a mouse, and provides a user with a function for setting imaging sequence conditions and image processing conditions. The display device 64 is a display device composed of, for example, a liquid crystal panel. The input device 62 and the display device 64 constitute a user interface.

  The image reconstruction unit 90 has a k-space database 92 therein. The image reconstruction unit 90 arranges raw data of MR signals input from the sequence controller 56 as k-space data in the k-space formed in the k-space database 92. The image reconstruction unit 90 performs image reconstruction processing including two-dimensional Fourier transform on k-space data to generate image data of each slice of the subject QQ. The image reconstruction unit 90 stores the generated image data in the image database 94.

  The image processing unit 96 takes in image data from the image database 94, performs predetermined image processing on the image data, and stores the image data after image processing in the storage device 66 as display image data.

  The storage device 66 stores the display image data, with the imaging sequence conditions used to generate the display image data, information on the subject QQ (patient information), and the like attached thereto as incidental information.

  The display control unit 98 causes the display device 64 to display an imaging sequence condition setting screen and an image indicated by image data generated by imaging under the control of the MPU 86.

  FIG. 3 is a circuit diagram showing an example of an equivalent circuit model of the gradient magnetic field generation system used in the calculation of the gradient magnetic field waveform. The gradient magnetic field generation system here refers to the entire components related to gradient magnetic field generation, such as the gradient magnetic field power supply 44, the gradient magnetic field coil 26, and the sequence controller 56 in FIG.

  Although the actual gradient magnetic field generation system of the MRI apparatus 20 is different from the configuration shown in FIG. 3, the gradient magnetic field calculation unit 306 has the circuit configuration of FIG. 3 for the gradient magnetic field generation system related to generation of the X-axis gradient magnetic field Gx. Assuming that the gradient magnetic field waveform is calculated.

  As shown in FIG. 3, the equivalent circuit model 140x includes, as a primary side, an X-axis gradient magnetic field power supply 44x, a resistor 26xR corresponding to a resistance component of the X-axis gradient magnetic field coil 26x, and an inductance of the X-axis gradient magnetic field coil 26x. The coil 26xL corresponding to the component is connected in series.

  The equivalent circuit model 140x has a series circuit of a resistor 141R and a coil 141L as a first secondary circuit. Further, the equivalent circuit model 140x includes a series circuit of a resistor 142R and a coil 142L as a second secondary circuit. The coil 26xL and the coil 141L are electromagnetically coupled to each other. Further, the coil 26xL and the coil 142L are electromagnetically coupled to each other.

  The meaning of the equivalent circuit model 140x having the above configuration will be described below. As the frequency increases, the impedances of the X, Y, and Z axis gradient magnetic field coils 26x, 26y, and 26z do not simply increase as in a simple model expressed by the sum of one resistance component and one inductance component. Absent.

  For example, in practice, when a high frequency current flows through a conductor, the current density is high on the surface of the conductor and decreases as it moves away from the surface. That is, due to the skin effect, the higher the frequency, the more the current is concentrated on the surface, so the AC resistance of the conductor becomes higher. In consideration of the skin effect and the like, it is desirable that the term including the resistance value of the resistor 26xR of the X-axis gradient magnetic field coil 26x in the impedance of the gradient magnetic field generation system expressed by a polynomial also varies depending on the frequency.

  Therefore, for example, an equivalent circuit model in which the resistance value of the resistor 26xR is multiplied by the angular frequency ω, that is, an equivalent circuit model in which the frequency dependence is reflected not only in the imaginary part but also in the real part of the impedance can be considered. desirable.

  In practice, when a pulse current is supplied to the X, Y, and Z-axis gradient magnetic field coils 26x, 26y, and 26z, an eddy current is generated, and the magnetic field due to the eddy current is added to each of the gradient magnetic fields Gx, Gy, and Gz, and the gradient magnetic field is generated. Distribution distortion occurs. Considering the magnetic field generated by the eddy current, it is desirable to consider an equivalent circuit model including mutual inductance.

  Although not shown in the figure, an actual gradient magnetic field generation system may include a choke coil that cuts off a high-frequency current of a predetermined frequency or higher. In this case, it is desirable to consider an equivalent circuit model including a plurality of mutual inductances instead of only one.

  The equivalent circuit model 140x shown in FIG. 3 is an example of an equivalent circuit model considering the above, and the equivalent circuit model is not limited to the configuration of FIG. For example, the number of coils other than the gradient coil 26 in the equivalent circuit is not limited to 2, and may be 1 or 3 or more.

  Further, as described above, the series circuit of the resistor 141R and the coil 141L and the series circuit of the resistor 142R and the coil 142L correspond to mutual inductance components and resistances included in the magnetic field generated by the eddy current and the skin effect. It is a component that actually exists.

  In this specification, for convenience, a series circuit of the resistor 141R and the coil 141L is regarded as a “virtual coil” that is a coil different from the gradient magnetic field coil 26. For this reason, in this specification, for convenience, the resistance value of the resistor 141R is regarded as the resistance component of the virtual coil, and the self-inductance value of the coil 141L is regarded as the self-inductance value of the virtual coil.

  However, the series circuit of the resistor 141R and the coil 141L is not a component that does not actually exist, but represents a mutual inductance component that actually exists in an equivalent circuit. The above interpretation also applies to the series circuit of the resistor 142R and the coil 142L in FIG.

Hereinafter, a method of calculating the gradient magnetic field waveform by the gradient magnetic field calculation unit 306 will be described.
In FIG. 3, resistance values of resistors 26xR, 141R, and 142R are Rload, R ₁ , and R ₂ , respectively. The self-inductance values of the coils 26xL, 141L, and 142L are Lload, L ₁ , and L ₂ , respectively. Further, a coil 26XL, the mutual inductance of the coil 141L and _{M 1.} Further, to the coil 26XL, the mutual inductance of the coil 142L and _{M 2.}

Further, the current value flowing in the direction of the arrow in FIG. 3 in the primary circuit is Iout (t). Further, the current value flowing in the direction of the arrow in FIG. 3 in the first secondary circuit is I ₁ (t). Further, the current value flowing in the direction of the arrow in FIG. 3 in the second secondary circuit is I ₂ (t). Further, the voltage value at both ends of the coil 26xL is set to Vout (t) with the arrow direction in FIG. 3 as the positive direction.

  (T) included in these codes means a function of time t, and the same applies to other codes used in the following description. At this time, the following expressions (1), (2), and (3) are established for the primary side and the secondary side, respectively.

In the formulas (1) to (3), M ₁ is represented by the following formula (4), and M ₂ is represented by the following formula (5).

(4) _{K 1} in the formula is a coupling coefficient between coils 26xL and the coil 141L, a coupling coefficient between the _{K 2} coil 26xL the coil 142L in (5). Here, the unit of the imaginary number is represented by j. That is, the square of j is -1.

  In the case of alternating current, the following equation (6) can be obtained by changing the equation (2) by replacing the time derivative d / dt with j × ω, and the following equation (7) can be obtained by modifying the equation (3). can get.

  In the equivalent circuit model 140x, Z represents the impedance of the X-axis gradient magnetic field coil 26x viewed from the X-axis gradient magnetic field power supply 44x. By dividing both sides of equation (1) by Iout (t), replacing time differentiation d / dt with j × ω, and substituting equations (6) and (7) into equation (1), impedance Z Is represented by the following equation (8).

  From the equation (8), the real part Re {Z} and the imaginary part Im {Z} of the impedance Z viewed from the X-axis gradient magnetic field power supply 44x are expressed by the following equations (9) and (10), respectively.

  The circuit constants A, B, C, and D in the expressions (9) and (10) are expressed by the following expressions (11), (12), (13), and (14), respectively.

  FIG. 4 is a graph schematically showing measured values of the frequency characteristics of the real part Re {Z} of the impedance Z of the X-axis gradient magnetic field coil 26x. In FIG. 4, the horizontal axis indicates the frequency, and the vertical axis indicates the real part Re {Z} of the impedance Z.

  FIG. 5 is a graph schematically showing measured values of frequency characteristics of Im {Z} / ω obtained by dividing the imaginary part Im {Z} of the impedance Z of the X-axis gradient magnetic field coil 26x by ω. In FIG. 5, the horizontal axis indicates the frequency, and the vertical axis indicates Im {Z} / ω.

  For example, the resistance value Rload in the equation (9) is determined in advance by measuring the voltage across the X-axis gradient magnetic field coil 26x when a direct current is passed through the X-axis gradient magnetic field coil 26x, and stored as a circuit constant. Stored in the unit 308 in advance. Here, for the measurement, for example, an LCR meter may be used (LCR of LCR is Inductance, C is Capacitance, and R is Resistance).

  Further, the self-inductance value Lload in the equation (10) can be calculated by measuring the magnetic flux generated by the X-axis gradient magnetic field coil 26x when a direct current is passed through the X-axis gradient magnetic field coil 26x, for example.

  Alternatively, the self-inductance value Lload may be determined by measuring with an LCR meter at a low frequency (for example, 1 to 10 hertz) that is not affected by the secondary side in the equivalent circuit model 140x.

Alternatively, as the self-inductance value Lload, a theoretical value may be calculated based on the shape (coil winding), material, and the like of the X-axis gradient magnetic field coil 26x and used.
The self-inductance value Lload determined in advance in this way is stored in advance in the circuit constant storage unit 308.

If the resistance value Rload and the self-inductance value Lload are determined as described above, the coupling coefficient K _{1 in the} equation (4), the coupling coefficient K _{2 in the} equation (5), and the time constant τ ₁ = L ₁ / R ₁ The time constant τ ₂ = L ₂ / R ₂ can be determined by the following fitting, for example.

Specifically, the frequency characteristic of the real part Re {Z} of the impedance Z expressed by the equation (9) and the measured value of the frequency characteristic of the real part of the impedance of the X-axis gradient magnetic field coil 26x are fitted.
Further, the frequency characteristic of Im {Z} / ω obtained by dividing the imaginary part Im {Z} of the impedance Z represented by the equation (10) by the angular frequency ω, and the imaginary part of the impedance of the X-axis gradient magnetic field coil 26x The measured value of the frequency characteristic divided by the frequency ω is fitted.

  In the above fitting, the real part Re {Z} and Im {Z} / ω obtained by dividing the imaginary part by the angular frequency ω are used, but the impedance is the sum of the expressions (9) and (10). The calculated value and measured value of Z (amplitude and phase thereof) may be used for fitting.

Alternatively, the calculated value and the measured value of the phase difference between the real part Re {Z} and the imaginary part Im {Z} of the impedance Z may be used for fitting.
If the coupling coefficient K ₁ and the coupling coefficient K ₂ , the time constant τ ₁ and the time constant τ ₂ are determined, the mutual inductances M ₁ and M ₂ are determined by the expressions (4) and (5), and (11) to (11) The circuit constants A, B, C, and D are determined by the equation (14).

  The circuit constants Rload, Lload, A, B, C, and D obtained as described above are stored in advance in the circuit constant storage unit 308. Thereby, the real part Re {Z} and the imaginary part Im {Z} of the impedance Z at an arbitrary frequency can be calculated by the equations (9) and (10).

  Here, in FIG. 3, the current sensitivity of the coil 26xL which is the main coil is α, the current sensitivity of the coil 141L is β, and the current sensitivity of the coil 142L is γ. The current sensitivity is a constant obtained by dividing the gradient magnetic field strength (Tesla / meter) generated by passing a current through the coil by the current value (ampere) passed through the coil. In this case, the X-axis gradient magnetic field waveform Gx (t) obtained by adding magnetic fields such as eddy currents can be calculated as a combined magnetic field waveform as shown in the following equation.

Gx (t) = α × Iout (t) + β × I ₁ (t) + γ × I ₂ (t) (15)

  The second and third terms on the right side of equation (15) are examples of magnetic field waveforms (virtual magnetic field waveforms) generated by the coils 141L and 142L.

As described above, the output current Iout (t) flowing through the X-axis gradient magnetic field coil 26x is determined by the conditions of the imaging sequence. Therefore, if the initial value of the current I ₁ (t) and the initial value of the current I ₂ (t) are determined in the expressions (2) and (3), the current flows in the secondary coil 141L in the expression (15). The current I ₁ (t) and the current I ₂ (t) flowing through the coil 142L can be determined based on the equations (6), (7) and each circuit constant. The initial value of the current I ₁ (t) and the initial value of the current I ₂ (t) are both set to zero, for example, assuming that a sufficient time has elapsed since the previous imaging sequence was executed. Can do.

Then, there is no unknown in equation (15), and the X-axis gradient magnetic field waveform Gx (t) can be calculated. Further, the Y-axis gradient magnetic field waveform Gy (t) and the Z-axis gradient magnetic field waveform Gz (t) can also be calculated based on the circuit model similar to FIG. 3 and the above equations (1) to (15).
As described above, the gradient magnetic fields Gss, Gpe, and Gro in the slice selection direction, the phase encoding direction, and the readout direction are formed by X-axis, Y-axis, and Z-axis gradient magnetic fields Gx, Gy, and Gz (combination). The Therefore, if the X-axis gradient magnetic field waveform Gx (t), the Y-axis gradient magnetic field waveform Gy (t), and the Z-axis gradient magnetic field waveform Gz (t) can be calculated as described above, the slice selection direction, the phase encoding direction, and The waveforms of the gradient magnetic fields Gss, Gpe, and Gro in the readout direction can also be calculated.

In the first embodiment, as an example, the accuracy of regridding processing for reconstruction is improved by changing the interval of reception sampling based on the X-axis gradient magnetic field waveform Gx (t) expressed by equation (15). To do. More specific description will be given below.
As described above, the gradient magnetic field is generated by applying a pulsed current to the gradient magnetic field coil 26. Although the shape of the current pulse applied to the gradient magnetic field coil 26 is ideally a rectangular wave, it is actually a trapezoidal wave having a rising region and a falling region. As a result, the pulse waveform of the gradient magnetic field itself is not an ideal rectangular wave but a trapezoidal wave having a rising region and a falling region.

  On the other hand, in a high-speed imaging method such as EPI, data is used as sampling (Ramp Sampling) and image reconstruction data not only in a flat pulse area but also in a rising and falling area. Thereby, the data collection period can be further shortened.

  MR signals sampled at regular intervals in the rising and falling regions are sampled when the gradient magnetic field is changing, and therefore are not equally spaced in the k space. Therefore, it is desired that the sampled MR signals be rearranged before image reconstruction so that they are equally spaced in the k space. This rearrangement process is a regridding process.

  FIG. 6 is a schematic diagram showing the concept of conventional regridding processing. In FIG. 6, each horizontal axis indicates the elapsed time t. Here, as an example, it is assumed that the X axis of the apparatus coordinate system coincides with the reading direction, and the gradient magnetic field current supplied to the gradient magnetic field coil 26x is Iout (t) (the same applies to the description of the next FIG. 7).

  Each of the upper left and lower left of FIG. 6 shows an example of the waveform of the gradient magnetic field current Iout (t), and the vertical axis thereof shows the magnitude of the gradient magnetic field current Iout (t). Each of the upper right and lower right of FIG. 6 shows an example of the waveform of the readout direction gradient magnetic field Gro, and the vertical axis thereof shows the magnetic field strength.

  In the prior art, the gradient magnetic field waveform was estimated on the assumption that the waveform of the readout direction gradient magnetic field Gro is similar to the waveform of the gradient magnetic field current Iout (t). That is, in the prior art, it has been estimated that the waveform of the readout direction gradient magnetic field Gro shown in the upper right of FIG. 6 can be obtained from the gradient magnetic field current Iout (t) in the upper left of FIG. Similarly, in the prior art, it has been estimated that the waveform of the readout direction gradient magnetic field Gro shown in the lower right of FIG. 6 can be obtained from the lower left gradient magnetic field current Iout (t) of FIG.

  As described above, in the prior art, the regridding process is performed based on the inaccurately estimated gradient magnetic field waveform.

  The gradient magnetic field current waveform may be trapezoidal as shown in the upper right of FIG. 6 or may contain a nonlinear component as shown in the lower right of FIG. In any case, in the prior art, the regridding process is executed on the assumption that the gradient magnetic field waveform is similar to the waveform of the gradient magnetic field current.

  However, the waveform of the gradient magnetic field current does not necessarily match the gradient magnetic field waveform actually generated from this gradient magnetic field current. In particular, in a waveform with a high frequency component, such as a gradient magnetic field used in high-speed imaging methods such as EPI, the divergence between the gradient magnetic field current waveform and the gradient magnetic field waveform increases, and the gradient magnetic field rises and falls. The discrepancy becomes significant.

  In the MRI apparatus 20 of the first embodiment, in order to cope with this problem, an actually generated gradient magnetic field waveform is accurately obtained under the assumption that there are a plurality of “different coils” from the gradient coil 26. calculate.

FIG. 7 is a schematic diagram illustrating a concept of a gradient magnetic field waveform calculation method according to the first embodiment. Processing proceeds in the order of upper left, lower left, lower right, and upper right in FIG. The upper left of FIG. 7 shows an example of the waveform of the gradient magnetic field current Iout (t) in the same manner as FIG. The lower right of FIG. 7 is a schematic diagram showing an example of waveforms of currents I ₁ (t) and I ₂ (t) flowing through the coils 141L and 142L of FIG. The upper right of FIG. 7 is a schematic diagram showing an example of a waveform of the calculated readout direction gradient magnetic field Gro.

In the first embodiment, when the gradient magnetic field current Iout (t) is input to an equivalent circuit including the gradient coil 26 and “another coil 141L, 142L”, the “another coil 141L, 142L” The flowing currents I ₁ (t) and I ₂ (t) are calculated (lower right in FIG. 7).

Based on the gradient magnetic field current Iout (t) and the calculated currents I ₁ (t) and I ₂ (t), the waveform of the actually generated readout direction gradient magnetic field Gro is calculated with high accuracy ( (Upper right of FIG. 7). Further, the regridding process is executed based on the gradient magnetic field waveform calculated with high accuracy in this way.

  The resistance value Rload in the equation (9) described in the first embodiment can be measured as described above, and the measured resistance value Rload is stored in the circuit constant storage unit 308 in advance. Further, the self-inductance value Lload determined in advance similarly to the first embodiment is stored in advance in the circuit constant storage unit 308.

If the resistance value Rload and the self-inductance value Lload are determined, the coupling coefficient K _{1 in the} equation (4), the coupling coefficient K _{2 in the} equation (5), the time constant τ ₁ = L ₁ / R _1, and the time constant τ ₂ = a _L 2 / _{R 2,} can be determined as in the first embodiment. Thereby, the mutual inductances M ₁ and M ₂ are determined in the same manner as in the first embodiment, and the circuit constants A, B, C, and D are determined by the equations (11) to (14).

  In the first embodiment, as an example, an example in which a gradient magnetic field waveform is calculated in a spin echo EPI and a regridding process is executed based on the gradient magnetic field waveform will be described.

FIG. 8 is a schematic diagram illustrating an example of a spin echo EPI imaging sequence.
FIG. 8 shows an example of waveforms of an RF pulse, a slice selection direction gradient magnetic field Gss, a phase encoding direction gradient magnetic field Gpe, a readout direction gradient magnetic field Gro, and an MR signal (echo signal) from the subject QQ in order from the top. . Each horizontal axis indicates the elapsed time t.

  EPI continuously inverts a gradient magnetic field at a high speed for one nuclear magnetic excitation, continuously generates MR signals, and collects these MR signals. More specifically, in EPI, for example, after application of a 90 ° excitation pulse, before the magnetization in the xy plane is attenuated by lateral relaxation (T2 relaxation) and disappears, the readout gradient magnetic field Gro is adjusted in accordance with the phase encoding step. Is reversed at high speed. Thereby, continuous gradient echoes are generated, and MR signals used for image reconstruction are collected.

  The EPI includes SE (spin echo) EPI, FE (field echo) EPI, FFE (Fast FE) EPI, and the like.

SE EPI is a technique using a spin echo method as shown in FIG. 8, and is a technique for collecting MR signals generated after a 90 ° excitation pulse and a 180 ° excitation pulse.
FE EPI is a technique for collecting MR signals generated after application of a 90 ° excitation pulse. FFE EPI is a technique that uses the fast FE method.

  Further, EPI that creates image data for one sheet by combining echo train data obtained by applying excitation pulses over a plurality of times is called multi-shot EPI. On the other hand, EPI that reconstructs an image only with echo train data obtained by applying a single excitation pulse is called single shot (SS) EPI.

Next, a configuration example of k-space data to be subjected to the gridling process will be described.
FIG. 9 is a schematic diagram illustrating an example of MR signal data immediately before being converted into k-space data when the number of matrix elements for phase encoding and frequency encoding is 256 × 256. In FIG. 9, TR is a repetition time, Ts in the horizontal direction is a sampling time, and the vertical direction is a phase encoding step (Phase Encode Step).

In this case, in principle, 256 lines of MR signals collected by changing the phase encoding 256 times are subtracted from the cosine function or sine function of the carrier frequency, respectively, and thereafter, for each phase encoding step as shown in FIG. Are lined up.
Here, for each ΔTs obtained by dividing the sampling time Ts of each MR signal in the horizontal direction of FIG. 9 by 256 at equal intervals, the intensity of the MR signal is set to a matrix value of each matrix element.

  As a result, 256 rows and 256 columns of matrix data are obtained for each of the real part (the one from which the cosine function is subtracted) and the imaginary part (the one from which the sine function is subtracted). These two matrix data are respectively referred to as k-space data.

  However, for example, in single shot EPI (echo planar imaging), if only 5 lines can be collected until the effective echo time, the number of collections is (256/2) + 5 = 133 lines. In this case, 123 lines that have not been collected contain, for example, zero as data in the k space.

  Here, the MR signals sampled at equal intervals or non-equal intervals under the application of the read direction gradient magnetic field Gro are the amounts obtained by integrating the read direction gradient magnetic field Gro in the time axis direction on the k space, that is, the read direction. This corresponds to the zeroth moment of the gradient magnetic field Gro.

  In the region where the gradient magnetic field is flat, the zero-order moment changes linearly. However, the zero-order moment in a region where the gradient magnetic field is not flat (such as a rising region and a falling region) is nonlinear. In high-speed imaging methods such as EPI, MR signal data sampled (Ramp Sampling) in regions where the gradient magnetic field is not flat (such as rising and falling regions) is also used for image reconstruction in order to further reduce the imaging time. Used.

  Image reconstruction assumes that the sampled data is in a linear region on k-space. For this reason, it is desired to convert or correct the non-linear sampling data so as to be linear in the k space.

FIG. 10 is a conceptual diagram showing that MR signals sampled at equal time intervals in a region where the readout direction gradient magnetic field Gro is nonlinear are non-equal in the k space.
The upper part of FIG. 10 is the same as the upper right part of FIG. 7 and shows an example of the waveform of the read direction gradient magnetic field Gro accurately calculated by the method of the first embodiment. That is, in the upper part of FIG. 10, the horizontal axis indicates the elapsed time t from the application start time of the pulse of the readout direction gradient magnetic field Gro, and the vertical axis indicates the magnetic field strength of the readout direction gradient magnetic field Gro.

  The middle part of FIG. 10 is obtained by time-integrating the absolute value of the magnetic field strength of the readout direction gradient magnetic field Gro in the upper part of FIG. The start of the integration period is in common the application start time of the pulse of the readout direction gradient magnetic field Gro. Therefore, in the middle part of FIG. 10, the horizontal axis indicates the end time of the integration period, and the vertical axis indicates the time integral value of the absolute value of the magnetic field strength of the reading direction gradient magnetic field Gro, that is, the zeroth-order moment.

  The lower part of FIG. 10 is a schematic diagram of each sampling period for one phase encoding step MR signal (that is, one line MR signal) when sampling is performed at equal intervals. In the lower part of FIG. 10, the horizontal axis indicates the elapsed time t from the application start time of the pulse of the read direction gradient magnetic field Gro, and the vertical axis indicates the intensity of the MR signal, as in the upper part. In this example, the number of frequency encoding steps is 256, and 256 sampling periods SP1, SP2, SP3, SP4,... SP256 are set. That is, one line of MR signals is divided at equal intervals into 256 sampling periods SP1 to SP256, as indicated by a one-dot chain line in the vertical direction of FIG.

  Here, “region where the gradient magnetic field is not flat” means a region where the zero-order moment of the gradient magnetic field is nonlinear, and “region where the gradient magnetic field is flat” means a region where the zero-order moment of the gradient magnetic field is linear. To do.

  Therefore, as can be seen from the upper, middle, and lower stages of FIG. 10, the read direction gradient magnetic field Gro is generated when the MR signal is sampled at equal intervals in time including the flat area and the non-flat area. k-space data is non-uniformly spaced in k-space.

  The MR signal sampled under the application of the readout direction gradient magnetic field Gro corresponds to the 0th order moment of the readout direction gradient magnetic field Gro in the k space. This is because, as shown by the alternate long and short dash line, the intervals are not equal.

  In the following description, the simple term “0th moment” refers to the 0th moment of the read direction gradient magnetic field Gro.

  FIG. 11 is a schematic diagram illustrating the concept of the first method of the regridding process in the first embodiment. The upper part of FIG. 11 is the same as the upper part of FIG. The lower part of FIG. 11 is a schematic diagram of each sampling period for MR signals for one phase encoding step when sampling is performed at non-uniform intervals.

  In the first method, one line of MR signals is divided into 256 sampling periods SP 1 ′, SP 2 ′, SP 3 ′ to SP 256 ′ at non-equal intervals, as indicated by a one-dot chain line in the vertical direction in FIG.

  The middle stage of FIG. 11 shows how to determine the sampling intervals SP1 'to SP256' at non-uniform intervals. The middle stage of FIG. 11 shows the 0th-order moment similar to that of FIG. That is, a one-dot chain line in the horizontal direction is drawn so that the zeroth moment increases at equal intervals. A one-dot chain line in the vertical direction is drawn so as to pass through each intersection point of each one-dot chain line in the horizontal direction and a thick line indicating the zeroth-order moment.

  In the first method, each sampling period is set such that each time integration value of the pulse intensity of the reading direction gradient magnetic field Gro having the representative time of each sampling period SP1 ′ to SP256 ′ as the end of the integration period is equally spaced from each other. SP1 ′ to SP256 ′ are defined. The start of the integration period in the above “time integration value” is, for example, the application start time of the pulse of the readout direction gradient magnetic field Gro in common. Further, the “representative time” may be, for example, the end time of each of the sampling periods SP1 ′ to SP256 ′ or the center time.

  In the first method, k-space data is generated by sampling MR signals at non-uniform intervals in time according to the sampling period determined as described above. The matrix elements of the k-space data generated in this way are arranged at equal intervals in the k-space. The phrase “arranged at equal intervals in the k space” means that the values of the zero-order moments corresponding to the respective sampling periods are arranged at equal intervals as indicated by the one-dot chain line in the horizontal direction in the middle of FIG. Means that.

  In other words, in the first method, the MR signals are arranged at non-equal time intervals so that the zero-order moments at the collection times (reception times) of the portions corresponding to the respective sampling periods in the MR signal are equally spaced. Sampled.

  In the first embodiment, since the gradient magnetic field waveform can be calculated with high accuracy based on the equivalent circuit as described above, the non-uniform sampling timing of the first method in FIG. 11 can be determined with high accuracy.

FIG. 12 is a schematic diagram illustrating the concept of the second method of the gridding process in the first embodiment.
The upper part of FIG. 12 shows matrix values ME1, ME2, and ME3 of matrix elements for one line (one row) of k-space data generated by sampling at equal intervals in time as in the lower part of FIG. , ME4,... ME256.

  Here, as an example, since the number of frequency encoding steps is considered to be 256, the number of matrix elements for one line (one row) is also 256. Each matrix value ME1, ME2, ME3,... ME256 corresponds to each sampling period SP1, SP2, SP3,.

  In the second method, as shown in the upper part of FIG. 12, the k-space data is once generated by sampling MR signals at regular intervals in time. Thereafter, the k-space data is rearranged (converted) as follows to become new k-space data.

  That is, rearrangement is performed so that the time integral values of the pulse intensities of the read direction gradient magnetic field Gro up to the representative time of the sampling period corresponding to each matrix element of the k-space data are equally spaced. For the rearrangement, a process such as interpolation may be used, and the “representative time” is the same as in the first method.

  The middle part of FIG. 12 shows the matrix values ME1 ′, ME2 ′, ME3 ′,... ME256 ′ of the k-space data after rearrangement on the upper side, and the original MR signal on the lower side (the same as the upper part of FIG. 12). ). These matrix values ME1 'to ME256' are values that should have been generated from the signal intensities of the respective sampling periods SP1 'to SP256' (same as in FIG. 11) in the original MR signal.

  The lower part of FIG. 12 shows the same 0th-order moment as the middle part of FIG. As shown by the one-dot chain line in the horizontal direction that divides the zeroth-order moment, which is the vertical axis of the lower stage of FIG. 12, at equal intervals, the zeroth-order moment corresponding to each matrix element of the k-space data after rearrangement is Become.

  That is, in the second method, the k-space data is rearranged so that the zero-order moments at the respective collection times (reception times) of the portions corresponding to the matrix elements in the MR signal are equally spaced. In the first embodiment, as described above, the gradient magnetic field waveform can be calculated with high accuracy based on the equivalent circuit. Therefore, in the second method of FIG. 12, k-space data is regenerated so as to obtain a uniform interval with high accuracy. Arrangeable.

  FIG. 13 is a flowchart illustrating an example of an operation flow of the MRI apparatus 20 according to the first embodiment. Hereinafter, the operation of the MRI apparatus 20 will be described according to the step numbers of the flowchart shown in FIG.

  [Step S21] The circuit constants in the equivalent circuit (see, for example, FIG. 3) are determined by the method described above. That is, the impedance of the gradient magnetic field coil 26 as viewed from the gradient magnetic field power supply 44 is measured, and circuit constants are calculated based on the measured values and the equations (9) and (10). Thereafter, the process proceeds to step S22.

  [Step S22] The circuit constant determined in step S21 is stored in the circuit constant storage unit 308. Note that the timing of executing the processes of steps S21 and S22 is not particularly limited as long as it is before the execution of the main scan.

  For example, it may be performed in a factory or the like before the product of the MRI apparatus 20 is shipped. Further, when the product is installed in a hospital or the like, it may be performed on-site. On-site implementation is effective when the surrounding environment of the factory and the surrounding environment of the installation site are significantly different. Further, when the surrounding environment of the installation place changes for some reason, the processes of steps S21 and S22 may be executed at an appropriate timing or periodically after the operation is started.

  Thereafter, the process proceeds to step S23. Steps S23 to S29 are processes when the main scan is executed.

  [Step S23] The gradient magnetic field calculation unit 306 reads each circuit constant of the equivalent circuit stored in the circuit constant storage unit 308. Thereafter, the process proceeds to step S24.

  [Step S24] The imaging condition setting unit 302 sets the imaging sequence of the main scan based on various types of information related to the imaging sequence conditions input by the user to the input device 62. The sequence controller control unit 300 inputs the imaging sequence set by the imaging condition setting unit 302 to the sequence controller 56. Thereafter, the process proceeds to step S25.

  [Step S25] The gradient magnetic field calculation unit 306 calculates a gradient magnetic field waveform based on the imaging sequence set in step S24 and the circuit constants of the equivalent circuit read in step S23. When the imaging sequence is set, the specifications of the gradient magnetic field (pulse train) in the imaging sequence are determined, and the waveform of the gradient magnetic field current Iout (t) to be passed through the gradient coil 26 is also determined. The gradient magnetic field calculation unit 306 calculates the gradient magnetic field waveform based on the waveform of the gradient magnetic field current and the circuit constant of the equivalent circuit.

  The gradient magnetic field waveform may be calculated for all three of the slice selection direction gradient magnetic field Gss, the phase encoding direction gradient magnetic field Gpe, and the readout direction gradient magnetic field Gro. In the first embodiment, at least the waveform of the readout direction gradient magnetic field Gro is calculated.

  More specifically, for example, when considering the equivalent circuit of FIG. 3, the X-axis gradient magnetic field waveform Gx (t) obtained by adding magnetic fields such as eddy currents can be calculated as a combined magnetic field waveform according to the equation (15).

  Here, as an example, if it is assumed that the X axis of the apparatus coordinate system matches the readout direction, the left side Gx (t) of equation (15) becomes the readout direction gradient magnetic field Gro (t). That is, the readout direction gradient magnetic field Gro (t) can be calculated by the equation (15).

As described above, the gradient magnetic field current Iout (t) flowing through the X-axis gradient magnetic field coil 26x is determined by the conditions of the imaging sequence. On the other hand, the circuit constants R ₁ , L ₁ , M ₁ , R ₂ , L ₂ , and M ₂ in the expressions (2) and (3) are already stored in the circuit constant storage unit 308. Therefore, if the initial value of the current I ₁ (t) and the initial value of the current I ₂ (t) are determined in the differential equations of the expressions (2) and (3), the current I ₁ (t) and the current I ₂ (t) time change (waveform) can be determined.

The initial values of the current I ₁ (t) and the current I ₂ (t) can both be zero, for example, assuming that sufficient time has passed since the previous imaging sequence was executed. Then, there is no unknown in equation (15), and the readout direction gradient magnetic field Gro (t) can be calculated. The gradient magnetic field waveform shown in the upper right of FIG. 7 is an example of the readout direction gradient magnetic field waveform Gro (t) calculated as described above. Thus, after the gradient magnetic field waveform including at least the readout direction gradient magnetic field Gro is calculated, the process proceeds to step S26.

  [Step S26] The main scan is executed in accordance with the imaging sequence of the main scan set in step S24, whereby MR signals are collected.

  More specifically, the subject P is placed on the top plate of the bed 32, and a static magnetic field is formed in the imaging space by the static magnetic field magnet 22 excited by the static magnetic field power supply 40. Further, a current is supplied from the shim coil power source 42 to the shim coil 24, and the static magnetic field formed in the imaging space is made uniform.

  When an imaging start instruction is input to the input device 62, an imaging sequence is input from the arithmetic device 60 to the sequence controller 56. The sequence controller 56 drives the gradient magnetic field power supply 44, the RF transmitter 46, and the RF receiver 48 according to the input imaging sequence, thereby forming a gradient magnetic field in the imaging region including the imaging part of the subject QQ. An RF pulse is generated from the RF coil 28 for transmission.

  Therefore, an MR signal generated by nuclear magnetic resonance inside the subject QQ is received by the receiving RF coil 28 and detected by the RF receiver 48. The RF receiver 48 performs predetermined signal processing on the detected MR signal, and A / D converts this to generate raw data that is a digitized MR signal. The RF receiver 48 inputs the raw data of the MR signal to the sequence controller 56. The sequence controller 56 inputs the raw data of the MR signal to the image reconstruction unit 90. Thereafter, the process proceeds to step S27.

  [Step S27] The regridding processing unit 320 performs a redding process based on the readout direction gradient magnetic field Gro (t) calculated in step S25 and the MR signal collected in step S26. The method of the gridding process here has already been described. That is, k-space data is generated based on the first method of FIG. 11, or k-space data once generated by sampling at equal intervals in time is rearranged according to the second method of FIG. The

  The k-space data subjected to the regridding process is stored in the k-space database 92. Thereafter, the process proceeds to step S28.

  [Step S <b> 28] The image reconstruction unit 90 performs image reconstruction processing on the k-space data after the regridding processing to generate image data, and stores the generated image data in the image database 94. Thereafter, the process proceeds to step S29.

[Step S29] When the imaging is not continued, the image processing unit 96 takes in the image data from the image database 94, performs predetermined image processing on the image data, generates two-dimensional display image data, and displays the display image. Data is stored in the storage device 66. Thereafter, the captured image of the main scan is displayed on the display device 64.
On the other hand, when imaging is continued, the arithmetic device 60 returns the process to step S23.
The above is the description of the operation of the first embodiment.

  As described above, the MRI apparatus 20 according to the first embodiment can accurately calculate the gradient magnetic field waveform based on the equivalent circuit model 140x close to the actual gradient magnetic field generation system in which the skin effect, eddy current, and the like are taken into consideration. Furthermore, in the first embodiment, the gradient magnetic field waveform can be calculated by the same method without spending a large calculation load on a desired imaging sequence. Since the regridding process is executed based on the gradient magnetic field waveform calculated accurately in this way, the accuracy of the regridding process can be improved.

  In the EPI pulse sequence, a slight waveform fluctuation in the rising and falling regions of the gradient magnetic field is very important in the regridding process. In the first embodiment, even when the EPI imaging sequence is slightly changed, the gradient magnetic field waveform reflecting the change can be immediately calculated, so that the adjustment processing time associated with the imaging sequence change can be reduced.

(Second Embodiment)
The second embodiment is a modification of the first embodiment.
FIG. 14 is a flowchart showing the operation of the MRI apparatus 20 of the second embodiment. The difference from the first embodiment is that a pre-scan (corresponding to steps S44 to S49) for verifying the circuit constants of the equivalent circuit is added. Hereinafter, the operation of the MRI apparatus 20 of the second embodiment will be described according to the step numbers shown in FIG.

  [Steps S41 to S43] The processing is the same as the processing in steps S21 to S23 of the first embodiment. Thereafter, the process proceeds to step S44.

  [Steps S44 to S48] In steps S44 to S48, image data of an evaluation image is generated as part of the pre-scan. Each processing content of steps S44 to S48 is the same as that of steps S24 to S28 of the main scan of the first embodiment, and the difference is that the generated image data is used as an evaluation image. Thereafter, the process proceeds to step S49.

  [Step S49] The constant correction unit 312 evaluates the image quality of the evaluation image based on, for example, the presence / absence of an artifact and the size of the artifact.

  In particular, in EPI, “N / 2 artifacts” due to variations between phase encodings are likely to occur. When the circuit constants of the equivalent circuit are not within the appropriate range, the calculation result of the gradient magnetic field waveform reflecting the relatively long period gradient magnetic field fluctuation that extends between the phase encodings is inaccurate.

  In this case, correction by the regridding process is also inaccurate, which may cause “N / 2 artifact”. In addition, even when there is a change in the surrounding environment such as a temperature change, the impedance of the gradient coil 26 when the circuit constant of the equivalent circuit is determined can change.

  Therefore, in step S49, the constant correction unit 312 automatically calculates the image quality reflecting the degree of artifact, for example, with a numerical evaluation index for the image data of the evaluation image automatically generated in step S48. The evaluation index here may indicate, for example, the degree of “N / 2 artifact”.

  When the evaluation index is less than the predetermined threshold, the constant correction unit 312 shifts the process to step S50. Otherwise, the process proceeds to step S51. Note that the case where the evaluation index is less than the predetermined threshold means that the artifact is equal to or higher than the predetermined level, and the image quality of the evaluation image is not sufficiently good. It should be noted that the quality of the evaluation image may be determined visually by a person, and the determination result may be input to the input device 62.

  [Step S50] The constant correction unit 312 performs a process of updating the circuit constant of the equivalent circuit, and then returns the process to step S44.

Specifically, for example, constant compensation unit 312, the value of circuit constants R ₁ of Figure 3 stored in the circuit constant storage unit 308 greatly updated, the process returns to step S44 again. Thereby, the evaluation image is generated again, and the evaluation index of the image quality is recalculated in step S49.

If the evaluation index becomes worse than the previously calculated value, constant compensation unit 312 re-update reduce the value of the circuit constants R ₁ stored in the circuit constant storage unit 308, the process returns to step S44 again. On the other hand, when the evaluation index is improved, the constant correction unit 312 keeps the value of the circuit constant R ₁ as it is, and greatly increases the value of other circuit constants such as L ₁ stored in the circuit constant storage unit 308. Update and return to step S44 again.

  Thus, for each parameter of the circuit constant, whether the value should be increased or decreased in order to improve the image quality, the circuit constant value is actually changed and the evaluation image is reproduced. Is determined. Such a pre-scan is repeated until the evaluation index of the image quality of the evaluation image reaches a predetermined threshold value.

  The improvement in image quality includes a reduction in artifacts. In other words, the pre-scan is repeated until the artifact is reduced to a predetermined level or less by updating the circuit constant by the constant correction unit 312. The above processing is merely an example of a circuit constant update method, and the circuit constant may be updated by another calculation method.

  [Step S51] Based on the latest circuit constant, that is, the circuit constant for which the artifact is determined to be equal to or lower than a predetermined level by the pre-scan, the MR signal acquisition and the regridding process of the main scan are executed. The processing content is the same as steps S23 to S28 of the first embodiment. The above is the description of the operation of the second embodiment.

  Thus, also in 2nd Embodiment, the effect similar to 1st Embodiment is acquired. Furthermore, in the second embodiment, the main scan is executed based on the circuit constants for which the artifact is determined to be equal to or lower than the predetermined level by the prescan. Therefore, it is possible to reliably acquire an image with good image quality.

(Third embodiment)
The third embodiment is different from the first embodiment in the use form of the calculated gradient magnetic field waveform. In the first embodiment, the regridding process is performed based on the calculated gradient magnetic field waveform. On the other hand, in the third embodiment, parameter correction processing related to the gradient magnetic field in the pulse sequence is performed based on the calculated gradient magnetic field waveform. Hereinafter, the parameter correction process will be described with reference to FIGS. 15 and 16.

  FIG. 15 is a schematic diagram illustrating an example of a parameter correction method for the slice selection direction gradient magnetic field Gss in EPI. In FIG. 15, each horizontal axis represents the elapsed time t, and each vertical axis represents the intensity of the slice selection direction gradient magnetic field Gss. The waveform of the slice selection direction gradient magnetic field Gss is set, calculated, and reset in the order of the waveform diagrams in the upper left, upper right, lower left, and lower right of FIG.

  As shown in the upper left of FIG. 15, in a normal slice selection direction gradient magnetic field Gss, a negative pulse called a rephasing lobe is applied immediately after a positive pulse.

  The initial values of parameters of this rephasing lobe (for example, application start timing, application end timing, magnetic field strength, etc.) are set as follows (as processing in step S64 in FIG. 17 described later).

That is, assuming that the positive-side pulse and the rephasing lobe are each a perfect trapezoidal wave, the area of the half of the positive-side pulse (S _A ) and the area of the rephasing lobe applied immediately thereafter (S _A ) _B ) is set to be equal. Here, “area” means, for example, the time integral value of the absolute value of the magnetic field strength.

  The waveform correction unit 316 determines the slice selection direction gradient magnetic field Gss based on the values of the parameters of the slice selection direction gradient magnetic field Gss set as described above and the above-described equivalent circuit and equations (1) to (15). Calculate the actual waveform.

The actual waveform of the calculated slice selection direction gradient magnetic field Gss is, for example, as shown in the upper right of FIG. In the calculated gradient magnetic field waveform, when the half-side area (S ′ _A ) of the positive pulse and the area of the rephasing lobe (S ′ _B ) are different (for example, when S ′ _A > S ′ _B ), The waveform correction unit 316 corrects the parameter value of the rephasing lobe.

For example, as shown in the lower left of FIG. 15, the waveform correction unit 316 extends the width of the rephasing lobe so that S ′ _A = S ′ _B. Then, the waveform correction unit 316 recalculates the waveform of the slice selection direction gradient magnetic field Gss based on the corrected parameter and the above-described equivalent circuit and equations (1) to (15).

The waveform correction unit 316 repeats correction of parameters such as the width of the rephasing lobe and calculation of the waveform of the slice selection direction gradient magnetic field Gss until S ′ _A = S ′ _B (lower right in FIG. 15).

FIG. 16 is a schematic diagram illustrating an example of a parameter correction method for the phase encoding direction gradient magnetic field Gpe in EPI. In FIG. 16, each horizontal axis indicates the elapsed time t. In the upper left, upper right, lower left, and lower right waveform diagram sets in FIG. 16, the vertical axis of the upper waveform diagram indicates the magnetic field strength of the readout direction gradient magnetic field Gro, and the vertical axis of the lower waveform diagram indicates the phase encoding direction gradient. The magnetic field strength of the magnetic field Gpe is shown.
The waveform of the phase encoding direction gradient magnetic field Gpe is set, calculated, and reset in the order of the waveform diagrams of the upper left, upper right, lower left, and lower right in FIG.

  First, on the assumption that the waveform of the readout direction gradient magnetic field Gro is a trapezoid, the rear end side (rear end portion) of the phase encode direction gradient magnetic field Gpe and the front end side (front end portion) of the readout direction gradient magnetic field Gro The initial values of the EPI pulse sequence parameters are set so as not to overlap in time (see the upper left of FIG. 16). This is a process in step S64 of FIG.

  Here, “overlapping in time” means, for example, that there is a timing at which both the gradient magnetic field pulse in the phase encoding direction and the gradient magnetic field pulse in the readout direction are applied.

  The waveform correction unit 316 calculates the phase encode direction gradient magnetic field Gpe and the read direction gradient magnetic field Gro based on the set values of such parameters and the above-described equivalent circuits and the equations (1) to (15). . The calculated gradient magnetic field waveform is, for example, as shown in the upper right of FIG.

  When the calculated rear end side of the pulse of the phase encoding direction gradient magnetic field Gpe and the front end side of the pulse of the readout direction gradient magnetic field Gro overlap in time, the waveform correction unit 316 sets the setting value of the phase encoding direction gradient magnetic field Gpe. Correct.

  Specifically, for example, the waveform correction unit 316 shortens the pulse width value of the phase encoding direction gradient magnetic field Gpe, and the amplitude (magnetic field strength) of the phase encoding direction gradient magnetic field Gpe so that the step width of the phase encoding does not change. Is increased (see the lower left in FIG. 16).

  The waveform correction unit 316 performs the phase encoding direction gradient magnetic field Gpe and the readout direction gradient based on the parameters of the EPI pulse sequence corrected as described above, the above-described equivalent circuit, and the equations (1) to (15). The magnetic field Gro is calculated again.

  The waveform correction unit 316 corrects the parameter values as described above until the rear end side of the pulse of the phase encoding direction gradient magnetic field Gpe and the front end side of the pulse of the readout direction gradient magnetic field Gro do not overlap in time. The calculation of the actual gradient magnetic field waveform based on the equivalent circuit is repeated.

  FIG. 17 is a flowchart showing an example of the operation of the MRI apparatus 20 according to the third embodiment. Hereinafter, the operation of the MRI apparatus 20 according to the third embodiment will be described according to the step numbers shown in FIG. 17 with reference to FIGS. 15 and 16 as appropriate.

  [Steps S61 to S64] The processing is the same as the processing in steps S21 to S24 in the first embodiment. Thereafter, the process proceeds to step S65.

  [Step S65] In steps S65 to S67, the waveform correction unit 316 executes parameter correction processing. In this step S65, the waveform correction unit 316 calculates the gradient magnetic field waveform as described above based on the EPI parameter value and the above-described equivalent circuit and equations (1) to (15) (FIG. 15). Upper right or lower right, see upper right or lower right in FIG. 16). Thereafter, the process proceeds to step S66.

  [Step S66] The waveform correction unit 316 determines whether or not the gradient magnetic field waveform calculated in step S65 sufficiently matches the target waveform.

  The above “sufficiently matches the target waveform” means that, for example, as described with reference to FIG. 15, in the slice selection direction gradient magnetic field Gss, the half area of the positive pulse and the area of the rephasing lobe are substantially equal. is there. Further, “sufficiently matches the target waveform” means that, as described with reference to FIG. 16 for example, the rear end side of the pulse of the phase encode direction gradient magnetic field Gpe and the front end side of the pulse of the read direction gradient magnetic field Gro It does not overlap.

  If it is determined that “the target waveform sufficiently matches”, the process proceeds to step S68, and if not, the process proceeds to step S67.

  [Step S67] The waveform correction unit 316 corrects the parameter setting value. Specifically, for example, in the slice selection direction gradient magnetic field Gss, when the area of the half of the positive pulse and the area of the rephasing lobe do not sufficiently match, the waveform correction unit 316 performs the rephasing so that both match. Extend the lobe width. The details are as described above with reference to FIG.

  When the rear end side of the pulse of the phase encoding direction gradient magnetic field Gpe and the front end side of the pulse of the readout direction gradient magnetic field Gro overlap temporally, the waveform correction unit 316 prevents the phase encoding direction gradient magnetic field from overlapping. The pulse width and amplitude (magnetic field strength) of Gpe are corrected. The details are as described above with reference to FIG.

  By such parameter correction processing, the gradient magnetic field waveform is brought close to the target waveform. After the parameter correction process described above, the process returns to step S65.

  [Step S68] When this step S68 is reached, the value of each parameter of the pulse sequence is initially set or corrected so as to sufficiently match the target waveform. The MRI apparatus 20 executes the main scan imaging sequence in the same manner as in step S11 of the first embodiment, and converts the collected MR signal data into k-space data and stores it. Thereafter, the process proceeds to step S69.

[Step S69] As in step S11 of the first embodiment, image reconstruction processing is performed on the k-space data, and display image data is generated and stored in the storage device 66. Thereafter, the display image data is transferred from the storage device 66 to the display control unit 98, and a captured image of the main scan is displayed on the display device 64.
The above is the description of the operation of the MRI apparatus 20 of the third embodiment.

  As described above, in the third embodiment, the gradient magnetic field waveform actually generated with the currently set parameters is calculated with high accuracy on the basis of the same equivalent circuit as in the first embodiment before the main scan is executed. Is done. If the calculated gradient magnetic field waveform does not sufficiently match the target waveform, such as when the rear end side of the phase encoding direction gradient magnetic field Gpe overlaps the front end side of the readout direction gradient magnetic field Gro in terms of time, before the execution of the main scan. The parameter value is corrected. Therefore, the gradient magnetic field waveform actually generated can be sufficiently matched with the target waveform, so that the image quality can be improved.

(Fourth embodiment)
The MRI apparatus 20 according to the fourth embodiment executes the regridding process according to the first embodiment and the parameter correction process according to the third embodiment.

  FIG. 18 is a flowchart showing the operation of the MRI apparatus 20 according to the fourth embodiment. The operation of the MRI apparatus 20 according to the fourth embodiment will be described below according to the step numbers shown in FIG.

  [Steps S81 to S88] The processing is the same as that of Steps S61 to S68 of the third embodiment. Thereafter, the process proceeds to step S89.

  [Steps S89 and S90] Steps S27 and S28 are the same as those in the first embodiment.

  As described above, in the fourth embodiment, the gradient magnetic field waveform is accurately calculated based on an equivalent circuit similar to that in the first embodiment. Then, the parameter correction process of the third embodiment is executed based on this gradient magnetic field waveform (step S87), and the regridding process of the first embodiment is executed after execution of the main scan (step S89). As a result, both the effects of the first embodiment and the effects of the third embodiment are obtained.

  As described above, according to the MRI apparatus 20 of the first to fourth embodiments, the actual gradient magnetic field waveform can be calculated with high accuracy, and therefore the regridding process and parameters are calculated based on the calculated gradient magnetic field waveform. The correction process can be executed with high accuracy.

(Supplementary items of the embodiment)
[1] In each of the above embodiments, the example in which the gradient magnetic field waveform is calculated based on the equivalent circuit model of the gradient magnetic field generation system shown in FIG. 3 has been described. The embodiment of the present invention is not limited to such an aspect.

  FIG. 19 is a circuit diagram showing another example of an equivalent circuit model of the gradient magnetic field generation system. An equivalent circuit model 140x ′ shown in FIG. 19 has a configuration in which a third secondary circuit is added to the equivalent circuit model 140x of FIG. The third secondary circuit has a configuration in which a capacitor 143C, a coil 143L, and a resistor 143R are connected in series. The coil 143L is electromagnetically coupled to the coil 26xL.

The capacitance value of the capacitor 143C is C ₃ , the self-inductance value of the coil 143L is L ₃ , the resistance value of the resistor 143R is R ₃ , and the current value flowing in the direction of the arrow in the third secondary circuit is I ₃ (t ), the mutual inductance between the coils 143L- coil 26xL and _{M 3.} At this time, since the following equations (16), (17), (18), and (19) are established, the gradient magnetic field waveform is based on these equations (16) to (19) as in the above embodiment. May be calculated.

  FIG. 20 is a circuit diagram showing still another example of an equivalent circuit model of the gradient magnetic field generation system. The equivalent circuit model 140x ″ of FIG. 20 is the same as the equivalent circuit model 140x ′ except that the connection of the capacitor 143C, the coil 143L, and the resistor 143R of the third secondary circuit is changed to a parallel connection.

In this case, in the third secondary circuit, the current value flowing through the coil 143L in the direction of the arrow in the figure is I ₃ (t), the current value flowing through the capacitor 143C in the same direction is I ₃₁ (t), and in the same direction A current value flowing through the resistor 143R is defined as I ₃₂ (t). As a result, I ₃ (t) is the sum of I ₃₁ (t) and I ₃₂ (t). Then, since the following formulas (20), (21), (22), (23), (24), and (25) are established, based on these formulas (20) to (25) A gradient magnetic field waveform may be calculated as in the above embodiment.

  [2] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

20 MRI device 26 Gradient magnetic field coil 30 Control device 56 Sequence controller 58 Computer 60 Arithmetic device 90 Image reconstruction unit 306 Gradient magnetic field calculation unit 312 Constant correction unit 316 Waveform correction unit 320 Regriding processing unit

Claims (12)

By applying a gradient magnetic field to the imaging region and sampling nuclear magnetic resonance signals collected from the imaging region, k-space data composed of a plurality of matrix elements is generated, and image data is based on the k-space data. A magnetic resonance imaging apparatus for reconstructing
A gradient magnetic field power source for applying the gradient magnetic field to the imaging region by flowing a gradient magnetic field current to the gradient coil according to the imaging sequence;
The gradient magnetic field waveform is calculated based on the conditions of the imaging sequence, and the gradient magnetic field waveform in the readout direction is calculated based on the mutual inductance in which the gradient coil generates mutual induction and the gradient magnetic field waveform. A gradient magnetic field calculating unit to
In the nuclear magnetic resonance signal, a portion where the time integral value of the gradient magnetic field strength in the readout direction is collected in a non-linear time zone is sampled, and up to the sampling period corresponding to each of the matrix elements. A magnetic resonance imaging apparatus comprising: a gridding processing unit that generates or rearranges the k-space data so that the time integration values are equally spaced. The regridding processing unit determines non-uniform sampling periods so that the time integration values having the representative time of each sampling period for the nuclear magnetic resonance signal as the end of the integration period are equally spaced from each other. The magnetic resonance imaging apparatus according to claim 1, wherein the k-space data is generated by performing sampling. The regridding processing unit generates the k-space data by sampling the nuclear magnetic resonance signal at regular intervals in time, and then represents each representative of the sampling period corresponding to each matrix element of the generated k-space data. The magnetic resonance imaging apparatus according to claim 1, wherein the k-space data is rearranged so that the time integration values having time as the end of the integration period are equally spaced from each other. The gradient magnetic field calculation unit includes a coil different from the gradient magnetic field coil, a gradient magnetic field coil, and an equivalent circuit of a gradient magnetic field generation system including the gradient magnetic field power supply. Inductance, self-inductance value and resistance value of the gradient magnetic field coil, self-inductance value and resistance value of the other coil are stored in advance as circuit constants, and based on the waveform of the gradient magnetic field current, the equivalent circuit, and the circuit constants The magnetic resonance imaging apparatus according to claim 1, wherein the gradient magnetic field waveform is calculated. The magnetic resonance imaging apparatus according to claim 4, wherein the gradient magnetic field calculation unit calculates the gradient magnetic field waveform based on the equivalent circuit including a plurality of the other coils. The magnetic resonance imaging apparatus according to claim 5, wherein the circuit constant is a constant determined in advance based on an actual measurement value of an impedance frequency characteristic of the gradient magnetic field coil. In the pre-scan before execution of the main scan, further comprising a constant correction unit for correcting the circuit constant,
The regridding processing unit generates or rearranges the k-space data of the prescan so that the time integration values are equally spaced with respect to the nuclear magnetic resonance signals collected in the prescan,
The constant correction unit calculates an evaluation index reflecting the degree of artifact for an evaluation image reconstructed based on the k-space data of the pre-scan, and calculates the circuit constant based on the evaluation index. The magnetic resonance imaging apparatus according to claim 4, wherein correction is performed. The imaging sequence is an imaging sequence of echo planar imaging,
The magnetic resonance imaging apparatus according to claim 7, wherein the artifact is an N / 2 artifact. When the gradient magnetic field waveform calculated by the gradient magnetic field calculation unit is different from the target waveform of the gradient magnetic field waveform, the waveform correction unit further includes a waveform correction unit that corrects the conditions of the imaging sequence before the execution of the imaging sequence. The magnetic resonance imaging apparatus according to claim 1. The imaging sequence is an imaging sequence of echo planar imaging,
The gradient magnetic field calculation unit further calculates a gradient magnetic field waveform in the phase encoding direction based on the mutual inductance and the waveform of the gradient magnetic field current,
When the rear end side of the gradient magnetic field waveform in the phase encoding direction calculated by the gradient magnetic field calculation unit and the front end side of the gradient magnetic field waveform in the readout direction overlap with each other in time, the waveform correction unit includes the rear end The magnetic resonance imaging apparatus according to claim 9, wherein the pulse width of the gradient magnetic field in the phase encoding direction is shortened and the intensity is increased so that the side and the front end side do not overlap. The magnetic resonance imaging apparatus according to claim 1, wherein the imaging sequence is an imaging sequence of echo planar imaging. A magnetic resonance imaging apparatus that applies a gradient magnetic field to an imaging region and reconstructs image data based on nuclear magnetic resonance signals collected from the imaging region,
A gradient magnetic field power source for applying the gradient magnetic field to the imaging region by flowing a gradient magnetic field current through the gradient coil according to an imaging sequence;
The gradient magnetic field waveform is calculated based on the conditions of the imaging sequence, and the gradient magnetic field waveform in the readout direction is calculated based on the mutual inductance in which the gradient coil generates mutual induction and the gradient magnetic field waveform. A gradient magnetic field calculating unit to
When the gradient magnetic field waveform calculated by the gradient magnetic field calculation unit is different from the target waveform of the gradient magnetic field waveform, a part of the conditions of the imaging sequence is set so that the gradient magnetic field waveform approaches the target waveform. A magnetic resonance imaging apparatus comprising: a waveform correction unit configured to perform correction before execution of the imaging sequence.

Images