• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    PURPOSE: A compensating MRI system for residual magnetization is provided an MRI system including a gradient compensation system which compensates imaging gradient waveforms for perturbations caused by residual magnetization. CONSTITUTION: An operation of a system is controlled from an operator console(100) which includes a keyboard and control panel(102) and a display(104). The console(100) communicates through a link(116) with a separate computer system(107) that enables an operator to control the production and display of images on the screen(104).The system control(122) includes a set of modules connected together by a backplane (118)including a CPU module(119) and a pulse generator module(121) which connects to the operator console(100) through a serial link(125). It is through this link(125) that the system control(122) receives commands from the operator which indicates the scan sequence that is to be performed. The pulse generator module(121) operates the system components to carry out the desired scan sequence. It produces data which indicates the timing, strength and shape of the RF pulses which are to be produced, and the timing of and length of the data acquisition window. The pulse generator module(121) connects through a gradient compensation system(129) to a set of gradient amplifiers(127), to indicate the timing and shape of the gradient pulses to be produced during the scan. The pulse generator module(121) also connects to a scan room interface circuit(133) which receives signals from various sensors associated with the condition of the patient and the magnet system. It is also through the scan room interface circuit 133 that a patient positioning system(134) receives commands to move the patient to the desired position for the scan. The gradient waveforms produced by the pulse generator module(121) are compensated by the system(129) as will be described in more detail below and applied to a gradient amplifier system(127) comprised of Gx, Gy and Gz amplifiers.
    公开号:KR20000047701A
    申请号:KR1019990052176
    申请日:1999-11-23
    公开日:2000-07-25
    发明作者:마징페이；조지아홍；미키논그라이미콜린；소버링제프리에스
    申请人:제이 엘. 차스킨, 버나드 스나이더, 아더엠. 킹；제너럴 일렉트릭 캄파니；
    IPC主号:
    专利说明:

    Residual magnetization compensation method and system for MR eye system {COMPENSATING AN MRI SYSTEM FOR RESIDUAL MAGNETIZATION}
    TECHNICAL FIELD The present invention relates to a method and apparatus for nuclear magnetic resonance imaging, and more particularly, to compensating for residual magnetization generated by a magnetic field in an MRI system.
    When a uniform magnetic field (polarization system B 0 ) is applied to a substance such as human tissue, each magnetic moment of the spin in the tissue tries to align with this polarization system, Proceed disorderly at the Larmor frequency. When the material or tissue is subjected to a magnetic field in the xy plane with an approximate lamor frequency, the net aligned magnetic moment (M z ) rotates and points toward the xy plane to produce a net cross magnetic moment (M t ). Create The signal is emitted by an excited spin, and after the excitation signal B 1 is terminated, the signal is received and processed to form an image.
    In order to apply magnetic resonance to imaging and many local spectroscopic techniques to selectively excite specific regions and to encode spatial information in the NMR system, the use of linear magnetic field gradients is necessary. During the NMR experiment, a magnetic field gradient waveform with a certain selected temporal change is used. Therefore, applying non-ideal magnetic field gradient waveforms, image distortion, intensity loss, ghosting, and other defects may be created. For example, if the partially selected magnetic field gradient is unbalanced before and after the 180 ° pulse, incomplete rephasing of the nucleus spindle, and hence signal loss, occurs. This results in the synthesis of later spin echoes in the Carr-Purcell-Mieboom-Gill sequence. Also, if the inclinometer is nonzero when it is certain (based on residual magnetization after the end of the inclination pulse), the distorted spectra in the Chemical Shift Imaging (CSI) sequence due to unwanted phase dispersion, And incorrect spin-spin relaxation time (T2) determination in multi-echo sequences. It is important to those skilled in the art that time varying magnetic field slopes are generated accurately.
    If the inclinometer combines conductive structures in a polarizing magnet, such as a cryostat, a shim coil system, or an RF shield used to separate the inclined coil from the RF coil (if the magnet is a superconducting device) One of the factors causing distortion in the generation of the slope may occur. The current induced in such an environment structure is called eddy current. Due to the eddy current, the magnetic field gradient increased exponentially before and after applying the ladder current pulse to the gradient coil, respectively.
    In the "field gradient eddy current compensation method" of U.S. Patent 4,698,591, an analog pre-emphasis filter is applied to a gradient power supply to form a current applied to the gradient coil in such a way that the eddy current induced gradient meter distortion is reduced. A method of use is disclosed. The filter includes a number of exponential decay components and an adjustable potentiometer and must be set during system calibration. Prior to system measurements, measurement techniques are used to calculate the potentiometer that measures the impulse response of the uncorrected magnetic field gradient and sets the pre-emphasis filter. Such techniques are described in US Pat. Nos. 4,950,994, 4,698,591, and 4,591,789.
    There are other types of gradient induced magnetic field perturbations in iron-core permanent magnets or iron core core reinforced superconducting magnets. This perturbation, known as hysteresis, has not been studied much, and the correction technique developed is not perfect. To understand the hysteresis phenomenon, reference is made to the effect of the bipolar gradient waveform shown in FIG. 2, and assume that the iron plate magnetization is in the initial state 8 as shown in FIG. This initial magnetization state is defined as a non-magnetization state, in which case it may be the state after the magnetic field ramps up and before any slope is applied. During the first initiation ramp, the current in the gradient coil, and the magnetic field H due to the iron plate core gradually increase. As a result, the magnetic induction B increases with H, as shown by the curve 11 in FIG. The ramp ramps down from 12 to 0, but magnetic induction B does not return to zero. Instead, they have different curve 14 characteristics regardless of the magnetic field. This phenomenon is called hysteresis, and the remaining magnetic induction (ΔB) is called residual magnetism. If the slope further ramps down to a negative value at 16, magnetic induction B proceeds along curve 18. In the subsequent ramp ramp 20, the H to B curve 22 ends with a negative residual magnetization (-ΔB). The candidate slope pulses drive magnetization in a loop called a hysteresis loop. It will be appreciated that the particular shape of the hysteresis loop depends on the configuration of the MRI system and typically consists of curved lines.
    The above analysis indicates that when a magnetic field gradient pulse over time is used for imaging, a perturbating magnetic field ΔB can be generated in the ferromagnetic material. If the hysteretic effect is not compensated for, a number of imaging artificial defects may be generated. For example, residual magnetization induced by phase-encoded gradient pulses may create contradictory phase errors in k-space data, resulting in image blur and ghosts.
    This issue is addressed in US Pat. No. 5,725,139. The solution proposed in this prior art is to correct the phase error produced by residual magnetization. Ten specific methods of doing this have been proposed, which require conversion to gradient pulse waveforms in certain specified pulse sequences.
    Accordingly, the present invention is a method and apparatus for reducing residual magnetization in an MRI system, thereby reducing artificial defects in an image. In particular, the present invention includes a method for calibrating an MRI system, in which the residual magnetization generated by the candidate imaging tilt waveform to be used in the MRI system is measured, and a compensation transformation for the candidate imaging tilt waveform is determined. If the candidate imaging tilt is used substantially by the MRI system to obtain the image data, a compensation transformation is made, and the resulting tilt does not produce residual magnetization. Compensation transformation includes attaching a compensation gradient lobe to the candidate imaging gradient waveform and converting the width of the candidate imaging gradient waveform.
    The residual magnetization can be divided into two parallel components, which can produce artificial defects in the MR image. The first is the stagnant hysteresis component, and the second is the transient hysteresis component. The calibration process measures the two hysteresis components and determines the compensation transform in the candidate imaging gradient waveform that reduces both effects.
    According to another aspect of the present invention, there is provided an effective method for compensating for imaging gradient waveforms. The calibration process can generate a look-up table of the compensation transform to be composed of the imaging tilt waveforms of various shapes. Appropriate compensation transformations may be composed of imaging gradients, which may then be used to obtain images directly from this lookup table, or compensation transformations may be computed through interpolation between the values stored in the lookup table. In another method, a polynomial that defines a compensation transformation as a function of the imaging gradient waveform shape can be determined during the calibration process. When an image is obtained, the compensation transformation for the imaging tilt can be calculated via this polynomial.
    1 is a block diagram of an MRI system using the present invention;
    FIG. 2 illustrates an exemplary imaging tilt waveform generated by the MRI system of FIG. 1;
    3 is a diagram illustrating a hysteresis curve generated by the imaging gradient waveform of FIG. 2;
    4 illustrates another exemplary imaging tilt waveform;
    5 is a diagram illustrating the hysteresis curve generated by FIG. 4;
    6 illustrates another exemplary imaging tilt waveform;
    7 is a diagram illustrating a hysteresis curve generated by the imaging tilt of FIG. 6;
    8 illustrates a pulse sequence executed by the MRI system of FIG. 1 to determine a reset slope for driving residual magnetization to zero;
    9 is a diagram illustrating a hysteresis curve generated by the pulse sequence of FIG. 8;
    10 illustrates a pulse sequence executed by the MRI system of FIG. 1 to generate a stagnant hysteresis compensated gradient lobe for an imaging gradient waveform;
    11 illustrates a pulse sequence executed by the MRI system of FIG. 1 adjusting the width of an imaging gradient waveform to compensate for transient hysteresis;
    12 is a flow chart for the calibration process performed by the MRI system of FIG. 1 to compensate for imaging gradient waveforms;
    13 is an electrical block diagram of a tilt compensation system forming the MRI system of part of FIG.
    Explanation of symbols for the main parts of the drawings
    100: operation console 102: control panel
    104: display 107: computer system
    108119: CPU 111: Disk Storage
    112: tape drive 113: memory
    115,125: serial link 118: back plan
    121: pulse generation module 122: separation system control
    127: slope amplification set 129: slope compensation system
    133: Scan Room Interface 134: Patient Placement System
    141: magnet assembly 150: transceiver
    151: RF amplifier 153: preamplifier
    154: T / R switch 161: array processor
    As shown in FIG. 4, if the oblique waveform 24 is consumed completely on the MRI system, the ferromagnetic structure of the MRI system will operate along the hysteresis curves 25 and 26 of FIG. 5. Residual magnetization DELTA B 1 is present at the completion of the slope waveform. Similarly, if negatively sloped waveforms 27 of the same magnitude are completely consumed as shown in FIG. 6, assuming that the initial magnetization state is not at the origin, magnetization follows the hysteresis curves 28 and 29 of FIG. It works. Residual magnetization (-ΔB 2 ) will be present at the completion of waveform 27. Residual magnetization (DELTA B 1 or-DELTA B 2 ) will produce a frequency offset from the Lamor frequency as follows.
    Δf 1 = ΔB 1 γ,
    Δf 2 = ΔB 2 γ.
    Here, γ represents a gyromagnetic ratio of spin. This frequency offset can be measured by generating the cross magnetization from the spin with the RF excitation pulse applied behind the slope waveforms 24 and 27 and also by the result of sampling the FID NMR signal. The Fourier transform of the obtained FID produces a peak at offset frequency Δf 1 or Δf 2 .
    If the hysteresis curves are perfectly symmetric, then Δf 1 is equal to -Δf 2 . Although this is not common, the offset frequency for zero magnetization can be computed perfectly from two measured frequencies as follows.
    Δf 0 = (Δf 1 -Δf 2 ) / 2
    This zero magnetization offset frequency is used in the next calibration procedure to create a slope "reset" that causes the residual magnetization to operate at zero.
    The reset slope waveform is determined by executing a pulse sequence on the MRI system shown in FIG. The sloped waveform is replaced at the anode, including an odd number of lobes 30 to 32 having a maximum magnitude. As shown in FIG. 9, this slope activates the magnetization of the ferromagnetic element from any residual magnetization shown at point 33, turns around the hysteresis curve of the largest possible magnitude, and ends at the positive maximum residual magnetization of point 34. do. Negative reset inclined lobe 35 is applied to a magnitude that operates the residual magnetization to zero, as shown by dashed line 36.
    The magnitude of the negative lobe 35 required to reset the residual magnetization to zero is determined by iterative processing using the pulse sequence of FIG. Once the ramp waveform has been consumed by the selected reset lobe 35, a non-selective RF excitation pulse 37 is applied to generate the cross magnetization. During the subsequent acquisition window 39 an FID signal is obtained and its frequency is determined. When the frequency of the measured FID 38 becomes equal to the zero magnetization offset frequency Δf 0, the pulse sequence is repeated by reset lobes 35 of different magnitudes. This reset slope is used to operate the residual magnetization to zero in the calibration measurement described later. The perturbation of the magnetic field causing residual magnetization can be divided into two components: stagnant hysteresis component and transient hysteresis component. The stagnant component of the residual magnetization remains for a long time without depending on the actual time after the gradient waveform is completed. The first correction method described later operates the stagnant component to zero by attaching a compensation gradient lobe to the gradient waveform to process the residual component. The transient hysteresis component is processed by changing the second correction method described later, i.e., changing the slope waveform area and offsetting the result.
    The above-described measurements for determining stagnation hysteresis component-reset slopes are provided to determine the magnitude of the compensation gradient lobe for any given magnitude gradient pulse. With particular reference to FIG. 10, the gradient waveform to be compensated for is represented by waveform 45, which is a phase coded gradient waveform, for example. In this "candidate" waveform, in order to operate the initial residual magnetization state to zero, the reset slope waveform is preceded by a negative compensation slope lobe 45 is then placed. In the reset gradient waveform, after the positive maximum magnitude gradient lobe, a negative gradient lobe having the above-described reset gradient magnitude is disposed or vice versa.
    The magnitude of the compensating gradient lobe 47 is set to zero the amount of residual magnetization produced by the candidate gradient waveform. This magnitude is determined through an iterative process similar to the above-described process in which the RF excitation pulse 48 is applied and the frequency of the resulting FID signal is determined. If the frequency of the FID (49) to be equal to the zero magnetization offset frequency (f 0 △) measured at the front of the candidate gradient waveform 45 it is properly compensated for stagnant hysteresis.
    It will be appreciated that if the candidate slope waveform is a negative slope, the compensation slope lobe is positive in magnitude. If the candidate slope waveform has positive and negative slope lobes (eg slice select slope), the appropriate compensated slope lobe will be positive or negative. This necessitates testing the size values over an appropriate range where the iterative process necessarily includes the optimal size.
    If it is possible to measure the compensation gradient lobes for all the imaging gradients used in all the imaging pulse sequences, it is possible to first measure the compensation gradients requiring changes in the gradient lobe size and width and store them in a table. This process is performed as part of the MRI system calibration. If a particular pulse sequence is continually determined, the appropriate compensation gradient magnitude is searched in the stored table and appended to the imaging gradient waveform. There are other ways to apply a specific compensation lobe size to the polynomial that represents the relationship between the candidate slope lobe size and width and the optimal compensation slope lobe size. Therefore, at run time, the optimal compensation slope lobe size is computed using this polynomial function.
    Transient Hysteresis Component- Due to transient hysteresis, a phase error occurs in spin magnetization in the cross section while the candidate slope waveform is consumed and for a short time thereafter. As described above, after the stagnation residual magnetization is compensated, the transient hysteresis component is measured using the pulse sequence of FIG.
    In particular, referring to FIG. 11, the residual magnetization is operated to zero by applying the reset slope waveform 52 described above. Transient magnetization is generated by a non-selective 90 ° RF excitation pulse 53, and then the candidate slope waveform 54 is applied. As discussed above, the candidate slope waveform 54 is compensated for stagnant hysteresis by the slope lobe 55. A non-selective 180 ° RF echo pulse 56 is applied at the time TE / 2 after the excitation pulse 53, and the NMR echo signal 57 is obtained facing the read slope 58. The read slope is selected and the area is the same as the entire area of the lobes 54 and 55. The read slope size is kept to a minimum, minimizing its own hysteresis effect.
    Ideally, the echo signal 57 is precisely aligned with the peak value occurring at the echo time TE after the RF excitation pulse 53. When candidate slope waveforms 54 and 55 are applied, due to the transient hysteresis effect, the echo signal is shifted in time as shown at 60 or 61. This phase offset is corrected by changing the width of the candidate pulses as shown at 63. The pulse sequence of FIG. 11 is repeated in an iterative process of changing the width of the candidate waveform 54 until the center of the echo signal 57 is precisely aligned at the ideal echo time TE.
    This procedure can be used to adjust the width of all gradient waveforms used on an MRI scanner, or as described above, the system used to adjust each gradient waveform with a given magnitude and pulse width at run time. A table or polynomial can be created during calibration. This calibration procedure compensates for any residual eddy currents with short time constants.
    First, referring to FIG. 1, the main configuration of the preferred MRI system related to the present invention is shown. Operation of the system is controlled from an operation console 100 that includes a keyboard and control panel 102 and a display 104. Console 100 communicates with each computer system 107 via a link, and each computer system 100 enables an operator to control the product and display an image on screen 104. Computer system 107 includes a number of modules that each communicate with one another via a backplane. The plurality of modules includes an image processing module 106, a CPU module 108, and a memory module 113 known in the art as a frame buffer for storing image data arrays. The computer system 107 is connected in link with the disk storage medium 111 and the tape driver 112 for image data and program storage, and communicates with each system controller 122 via the high speed serial link 115.
    System controller 122 includes a set of modules connected to each other by backplane 118. These include a pulse generation module 121 connected to the operation console 100 via the CPU module 119 and the serial link 125. Via this link 125, the system controller 122 receives instructions indicating the scan sequence to be executed from the operator. The pulse control module 121 operates the system configuration to execute the desired scan sequence. The pulse control module 121 generates data representing timing, intensity, intensity and shape of an RF pulse to be generated, and timing and length of a data acquisition window. To indicate the timing and shape of the gradient pulses to be generated during the scan, the pulse control module 121 is connected to the gradient amplifier set 127 through the gradient compensation system 129. The pulse generation module 121 is connected to a scan room interface circuit 133 that receives signals from various sensors related to the conditions of the patient and the magnet system. The patient placement system 134 receives commands via the scan room interface circuitry 133 to move the patient to the desired location for the scan.
    The gradient waveform generated by the pulse generation module 121 is compensated by the system 129 as described in more detail below and is applied to the gradient amplifier system 127 consisting of Gx, Gy, and Gz amplifiers. Each gradient amplifier excites a corresponding gradient coil (not shown) that forms part of the magnet assembly 141. As is well known in the art, gradient coils generate linear magnetic field gradients for use in position coded acquisition signals. The magnet assembly 141 includes a polarizing magnet (not shown) and an RF coil (not shown) entire body. In a preferred embodiment, the polarizer is created by a permanent magnet and associated iron plate cores used to form and manage the polarizer, as described in "Magnet Assembly for MRI Devices" in US Pat. No. 5,652,517. . These elements are magnetized by the inclinometer and cause residual magnetization problems to be addressed by the present invention.
    In the system controller 122, the transceiver module 150 generates RF pulses. The RF pulse is amplified by the RF amplifier 151 and coupled to the RF coil of the magnet assembly 141 by the transmit / receive switch 154. The resulting signal emitted by the patient's excited nuclei is sensed by the same RF coil and coupled to the preamplifier 153 via a transmit / receive switch 154. The amplified NMR signal is demodulated at the receiver of the transceiver 150, filtered and digitized. Transmit / receive switch 154 is controlled by a signal from pulse generation module 121 to electrically connect RF amplifier 151 and RF coil during transmission and preamplifier 158 during receive mode. . In addition, the transmit / receive switch 154 enables each RF calibration to be used in the transmit or receive mode, as described in more detail below.
    The NMR signal picked up by the RF coil is digitized by the transceiver module 150 and passed to the memory module 160 of the system control 122. When the scan is completed and the entire data array is obtained from the memory module 160, the array processor 161 is operated to convert the data into the image data array. This image data is transported to the computer system 107 via the serial link 115 and stored in the disk memory 111. In response to the command received from the operating console 100, this image data is stored on the external driver 112 or processed once more by the image processor 106, and transported to the operating console 100 for display. Appear on 104.
    For more details on transceiver 150, see US Pat. Nos. 4,952,877 and 4,992,736.
    The MRI system is periodically calibrated using the procedure shown in FIG. For the measurements necessary for this procedure, a calibration phantom is placed in the MRI system. This virtual image supports a sample of MR active material along one tilt axis and spaced apart from the system isocenter. For example, the sample is 0.44 cc 0.5 M CuSO 4 doped water contained in an acrylic tube approximately 1 inch in diameter. The sample serves as a source for the NMR signal in the measurement pulse sequence used to calibrate the MRI system. Prior to performing the calibration procedure of the present invention, the MRI system is compensated to offset the eddy current effect using one of the well known methods.
    In particular, with reference to FIG. 12, in the residual magnetization calibration procedure, the first step is to measure the zero magnetization offset frequency, as shown in processing block 200. This step has been described above in relation to Equations 1-3, where the zero magnetization offset frequency represents the NMR signal frequency when no residual magnetization is present in the MRI system. The magnitude of the reset slope waveform is determined as shown in process block 202. As described above, this step uses the pulse sequence of FIG. 8 to determine the shape of the reset slope waveform that drives the residual magnetization of the MRI system to zero. A loop is provided so that each candidate imaging gradient waveform is compensated for the stagnant hysteresis effect and the transient hysteresis effect. With continued reference to FIG. 12, at processing block 204, the candidate slope waveform, including magnitude and pulse width, is selected from the table that stores the slope shape list.
    As shown by processing block 206, the pulse sequence of FIG. 10 is executed to measure the frequency offset imposed by the candidate gradient waveform. If the measured frequency offset is within the preselected range of zero magnetization offset, as determined at decision block 208, processing proceeds to the next step. If not, the compensating tilt lobe (FIG. 10) is adjusted as shown in processing block 210, and the frequency offset measurement is repeated. This step is repeated until the optimal compensation slope lobe is found in the candidate slope waveform. As shown by processing block 212, this optimal compensated warp lobe size is stored in a look-up table. During this calibration process, the entire net slope area is maintained by adjusting the width of the candidate slope waveform.
    In the next step shown in processing block 214, the echo signal time shift is measured using the pulse sequence of FIG. As described above, the peak position of the echo signal 57 is detected, and the offset from the echo time TE is determined by calculating the 1D Fourier transform and measuring the linear phase slope. As determined at decision block 216, if the offset is within the preset limit region from the echo time TE, the compensation is complete. Otherwise, the width of the candidate slope waveform is adjusted at processing block 218 and the measurement of offset time is repeated at processing block 214. The optimal adjustment finally obtained by this iterative process is stored in the lookup table as shown in process block 220.
    The process is repeated for all gradient waveforms to be used on the MRI system. In addition, the process is repeated for each tilt axis, x, y, z. Once all candidate slope waveforms are compensated as shown in decision block 222, the calibration process is complete.
    It should be noted that in the above described calibration process, many changes are possible. Rather than process each and every candidate slope waveform used in the MRI system, a sample slope waveform of the selected size and width will be processed. The compensation value stored in the lookup table can be used continuously to directly compensate the sample slope waveform when used in the imaging pulse sequence. In addition, other oblique waveforms will be compensated using the values generated by interpolating between the values stored in the lookup table.
    Another method processes the sample slope waveform to generate a compensation set as a function of candidate slope waveform size and width. This sample compensation value is applied to a polynomial function that represents the compensation value as a function of gradient magnitude and width.
    When an image is acquired using a calibrated MRI system, the operator enters a predetermined scan into the MRI system. This rule defines a scan parameter that includes a particular gradient waveform to be used during the imaging pulse sequence. However, before being used, using the compensation value calculated during the gradient waveform calibration process, this gradient waveform is compensated for residual magnetization. Such compensation can be performed in a number of ways.
    In a preferred embodiment, some imaging oblique waveforms are compensated before scanning starts. These include slice select gradient waveforms and read slope waveforms that remain fixed during the scan. Compensation values for these gradient waveforms are read from the lookup table and used to modify their shape. This includes adding the compensation tilt lobe and the imaging tilt waveform, and adjusting the width of the imaging tilt waveform. The compensated gradient waveforms are then stored in pulse generation module 121 for use in generating imaging gradients during the scan in which they are all consumed.
    Some imaging tilts are compensated when they are all consumed by the pulse generating module 121. The phase coded slope in the imaging pulse sequence is stepped through a number of values (e.g. 128, or 256) during the scan, and in a preferred embodiment, these are gradient compensation systems (FIG. 1) when they are all consumed. Of 129).
    In particular, referring to FIG. 13, the tilt compensation system 129 includes a waveform memory 250 that stores, in digital form, a lookup table calculated during a calibration process. This stored lookup table is used to calculate the compensation waveform appended to the imaging gradient waveform when the controller 202 receives a command from the pulse control module 121 via the control bus 204. Used by. The digital values of this compensation waveform are applied to one or more A / D converters 256-258 via the data bus 260. The controller 252 enables the appropriate D / A converters 256-258, writes the digitized compensation waveforms to them, and displays an analog version of the compensation waveform at the output of one or more A / D converters 256-258. version). These outputs drive the x-axis, y-axis, and z-axis gradient amplifiers 127, respectively.
    When the imaging pulse sequence is executed by the MRI system of FIG. 1, the pulse generation module 121 generates imaging gradient waveforms on the data bus 260 and applies them to the appropriate D / A converters 256-258. Controller 252 is then signaled over control bus 254 to attach the compensation waveform. Controller 252 generates the appropriate compensation waveform and applies it to appropriate A / D converters 256-258.
    The completed slope calibration process involves the following steps:
    1. Compensate for eddy currents;
    2. Stagnant residual magnetization is zero while preserving the net slope region of the candidate slope waveform;
    3. Adjust the width of the candidate slope waveform to include a transient hysteresis effect and a transient eddy current of zero.
    As described above, the present invention has the effect of reducing the residual magnetization in the MRI system, thereby reducing the artificial defect of the image.
    权利要求:
    Claims (13)
    [1" claim-type="Currently amended] In the method of reducing the residual magnetization effect of the image obtained by calibrating the MRI system,
    a) selecting a candidate slope waveform 204 to be compensated for;
    b) measuring residual magnetization effects 206 and 214 produced by the selected candidate slope waveform with an MRI system;
    c) determining compensation adjustment data 210, 218 for the selected candidate slope waveform that significantly reduces the residual magnetization effect;
    d) repeating steps a), b), and c) for a plurality of different candidate gradient waveforms 222 used by the MRI system;
    e) storing the compensation adjustment data (212,220) for use by the MRI system when acquiring an MRI image.
    [2" claim-type="Currently amended] The method of claim 1,
    And the compensation adjustment data is stored as a compensation adjustment value in a lookup table.
    [3" claim-type="Currently amended] The method of claim 1,
    The compensation adjustment data is a residual magnetization compensation method for an MRI system stored in a polynomial representing the compensation adjustment as a function of the slope waveform shape.
    [4" claim-type="Currently amended] The method of claim 1,
    The step b) includes executing a pulse sequence with the MRI system using the selected candidate slope waveform 45 to measure the frequency shift of the NMR signal 49 due to the candidate slope waveform 45. Residual magnetization compensation method for MRI system.
    [5" claim-type="Currently amended] The method of claim 4, wherein
    The pulse sequence is,
    Iii) applying a reset ramp pulse waveform 46 to actually drive the residual magnetization of the MRI system to zero;
    Ii) applying the selected candidate slope waveform 45;
    Iii) generating an NMR signal by applying an RF excitation pulse 48;
    Iii) acquiring the NMR signal.
    [6" claim-type="Currently amended] The method of claim 4, wherein
    C),
    Attaching a compensating gradient lobe 47 to the candidate gradient waveform 45;
    Adjusting the size of the compensation gradient lobe 47, repeating step b) until the measured frequency shift is lower than a preset threshold,
    The resultant compensation slope lobe size is the compensation magnetization compensation data for the selected candidate slope waveform.
    [7" claim-type="Currently amended] The method of claim 4, wherein
    The step b) executes a second pulse sequence with the MRI system using the selected candidate slope waveform, so that the reverberation times of the NMR echo signals 57, 60, 61 caused by the candidate slope waveform 54 TE) Residual magnetization compensation method for MRI systems measuring shift.
    [8" claim-type="Currently amended] The method of claim 7, wherein
    The second pulse sequence is,
    Iii) applying a reset ramp pulse waveform to drive the residual magnetization of the MRI system to zero sufficiently;
    Ii) applying an RF excitation pulse 53 to generate a cross magnetization;
    Iii) applying the selected candidate slope waveform 54;
    Iii) applying an RF pulse 56 to invert the transverse magnetization and generate NMR echo signals 57, 60, 61;
    Iii) acquiring the NMR echo signal.
    [9" claim-type="Currently amended] The method of claim 7, wherein
    The step c) includes adjusting the width of the selected candidate slope waveform 54 and repeating step b) until the measured echo time TE is lower than a preset threshold,
    The resulting width adjustment is a residual magnetization compensation method for the MRI system that is compensation adjustment data for the selected candidate slope waveform.
    [10" claim-type="Currently amended] An MRI system comprising an inclination system 127 for generating an imaging magnetic field inclination during a scan in response to an imaging inclination waveform generated by the pulse generator 121,
    Tilt compensation system 129 coupled to the tilt system 127 and operable to attach a compensation tilt lobe to an imaging magnetic field waveform 450 that sufficiently activates the residual magnetization of the MRI system after the image magnetic field tilt is generated. Device with a.
    [11" claim-type="Currently amended] The method of claim 10,
    The slope compensation system (129) includes means for generating a reset slope generated during a scan to operate the residual magnetization of the MRI system to substantially zero.
    [12" claim-type="Currently amended] A method of generating a reset slope for an MRI system,
    a) executing a pulse sequence with the MRI system,
    Iii) generating a reset gradient waveform comprising a first lobe 32 and a second lobe 35;
    Ii) generating an RF excitation pulse 37 to generate transversal magnetization,
    Iii) obtaining an NMR signal 38;
    b) determining the frequency of the NMR signal;
    c) varying the size of the second reset warp lobe;
    d) repeating steps a) to c) until the frequency of the NMR signal is substantially equal to a zero magnetization frequency Δf 0 .
    [13" claim-type="Currently amended] The method of claim 12,
    The zero magnetization frequency Δf 0 executes a pulse sequence with the MRI system to measure the frequency Δf 2 of the NMR signal obtained after the gradient pulse of one pole is applied.
    Another pulse sequence is executed with the MRI system to measure the frequency Δf 2 of the NMR signal obtained after the gradient pulse of the opposite pole is applied,
    A method of calculating a zero magnetization frequency with Δf 0- (Δf 1 -Δf 2 ) / 2.
    类似技术:
    公开号 | 公开日 | 专利标题
    Barmet et al.2009|A transmit/receive system for magnetic field monitoring of in vivo MRI
    DE10041808B4|2011-07-28|Field frequency blocking system for a magnetic resonance system
    KR101070153B1|2011-10-05|Nuclear magnetic resonance imaging apparatus and nuclear magnetic resonance imaging method
    EP1582886B1|2012-08-15|Magnetic resonance apparatus with magnetic field detectors
    DE19905720B4|2010-10-14|Fast spin echo pulse train for diffusion-weighted imaging
    US4970457A|1990-11-13|MRI compensated for spurious rapid variations in static magnetic field during a single MRI sequence
    US6720767B2|2004-04-13|MRI apparatus
    US7015696B2|2006-03-21|Magnetic resonance imaging apparatus and magnetic resonance imaging method
    JP4201993B2|2008-12-24|Apparatus and system for measurement and compensation of eddy currents induced during NMR imaging operations
    KR100562003B1|2006-03-22|Coil sensitivity map generating method, parallel imaging method, and mri apparatus
    US6559642B2|2003-05-06|Calibration method for use with sensitivity encoding MRI acquisition
    US5151656A|1992-09-29|Correction of nmr data acquired by an echo-planar technique
    EP1328830B1|2010-01-13|Nmr sequence for optimizing instrument electrical power usage
    US5519320A|1996-05-21|Nuclear magnetic resonance imaging with high speed and interactive pulse sequence control
    US8698495B2|2014-04-15|Flip angle imaging with improved B1 mapping for multi-RF transmit systems
    Boesch et al.1991|Temporal and spatial analysis of fields generated by eddy currents in superconducting magnets: optimization of corrections and quantitative characterization of magnet/gradient systems
    DE19821780B4|2010-04-15|Correction of artifacts caused by Maxwell terms in cut-shift echo-planar imaging
    Brodsky et al.2009|Characterizing and correcting gradient errors in non‐cartesian imaging: Are gradient errors linear time‐invariant |?
    US6815952B1|2004-11-09|Magnetic resonance diffusion imaging with eddy-current compensation
    US5647362A|1997-07-15|Correction of read-gradient polarity in EPI and grase MRI
    US7689262B2|2010-03-30|Method and magnetic resonance tomography apparatus for correcting changes in the basic magnetic field
    JP4693209B2|2011-06-01|Fast spin echo phase correction method for magnetic resonance imaging system
    US5001428A|1991-03-19|Method for mapping the RF transmit and receive field in an NMR system
    EP2463677A1|2012-06-13|Gradient non-linearity compensation in moving table MRI
    DE10318990B4|2008-04-03|Imaging method for magnetic resonance tomography
    同族专利:
    公开号 | 公开日
    JP2000157509A|2000-06-13|
    CN1258001A|2000-06-28|
    EP1004892A1|2000-05-31|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    法律状态:
    1998-11-23|Priority to US19822698A
    1998-11-23|Priority to US09/198,226
    1999-11-23|Application filed by 제이 엘. 차스킨, 버나드 스나이더, 아더엠. 킹, 제너럴 일렉트릭 캄파니
    2000-07-25|Publication of KR20000047701A
    优先权:
    申请号 | 申请日 | 专利标题
    US19822698A| true| 1998-11-23|1998-11-23|
    US09/198,226|1998-11-23|
    [返回顶部]