WO1985005693A1 - Prise d'image de fourier nmr a partir d'echos multiples - Google Patents

Prise d'image de fourier nmr a partir d'echos multiples Download PDF

Info

Publication number
WO1985005693A1
WO1985005693A1 PCT/US1985/000950 US8500950W WO8505693A1 WO 1985005693 A1 WO1985005693 A1 WO 1985005693A1 US 8500950 W US8500950 W US 8500950W WO 8505693 A1 WO8505693 A1 WO 8505693A1
Authority
WO
WIPO (PCT)
Prior art keywords
applying
echoes
pulses
pulse
data set
Prior art date
Application number
PCT/US1985/000950
Other languages
English (en)
Inventor
Andrew A. Maudsley
Original Assignee
Advanced Nmr Systems, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Advanced Nmr Systems, Inc. filed Critical Advanced Nmr Systems, Inc.
Publication of WO1985005693A1 publication Critical patent/WO1985005693A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/561—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution by reduction of the scanning time, i.e. fast acquiring systems, e.g. using echo-planar pulse sequences
    • G01R33/5615—Echo train techniques involving acquiring plural, differently encoded, echo signals after one RF excitation, e.g. using gradient refocusing in echo planar imaging [EPI], RF refocusing in rapid acquisition with relaxation enhancement [RARE] or using both RF and gradient refocusing in gradient and spin echo imaging [GRASE]
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/561—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution by reduction of the scanning time, i.e. fast acquiring systems, e.g. using echo-planar pulse sequences
    • G01R33/5615—Echo train techniques involving acquiring plural, differently encoded, echo signals after one RF excitation, e.g. using gradient refocusing in echo planar imaging [EPI], RF refocusing in rapid acquisition with relaxation enhancement [RARE] or using both RF and gradient refocusing in gradient and spin echo imaging [GRASE]
    • G01R33/5617—Echo train techniques involving acquiring plural, differently encoded, echo signals after one RF excitation, e.g. using gradient refocusing in echo planar imaging [EPI], RF refocusing in rapid acquisition with relaxation enhancement [RARE] or using both RF and gradient refocusing in gradient and spin echo imaging [GRASE] using RF refocusing, e.g. RARE
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/565—Correction of image distortions, e.g. due to magnetic field inhomogeneities
    • G01R33/56563—Correction of image distortions, e.g. due to magnetic field inhomogeneities caused by a distortion of the main magnetic field B0, e.g. temporal variation of the magnitude or spatial inhomogeneity of B0

Definitions

  • the invention relates to techniques for obtaining an image from nuclear magnetic resonance (NMR) data. More specifically, the invention relates to techniques for applying pulses to an object to obtain a series of NMR echoes, which may then be processed and transformed, such as by a Fourier transform, to obtain an image.
  • NMR nuclear magnetic resonance
  • U.S. Patent No. 4,290,019, issued September 15, 1981 to J.M.S. Hutchison et al. discloses another technique for NMR Fourier imaging.
  • This technique includes the application of a. sequence of gradient fields and high frequency pulses to obtain a spin echo of the FID. After an echo is obtained, the atoms being imaged are permitted to relax before the sequence is repeated to obtain another echo. The echoes are processed to obtain an image, and the processing includes Fourier transformation.
  • the invention provides NMR Fourier imaging techniques with more sensitive data acquisition and increased image contrast between different types of tissue.
  • the invention is based on the discovery that sensitive data acquisition and increased image contrast can be obtained in NMR Fourier imaging by applying a pulse and gradient switching sequence which produces a series of echoes of each FID. These echoes can be combined into a data set which is then transformed into the image.
  • the invention is based on the discovery of a method of adding echoes together which preserves all phase information, so that some or all of a series of echoes may be combined without loss of amplitude or phase information.
  • the invention is based on the discovery of a pulse and gradient switching sequence which preserves the phase information despite errors which result from the application of non-ideal pulses.
  • a pulse and gradient switching sequence has been discovered which provides a series of FID echoes whose base line components cancel when added together, eliminating an undesirable artifact.
  • the method of the invention thus includes applying a sequence of stimulating signals, the pulse and gradient switching sequence, to the object to be imaged to produce a series of echoes.
  • the echoes are then combined to obtain a data set, and the data set is then transformed into an NMR image of the object.
  • the apparatus of the invention includes means for applying stimulating signals to an object, the applying means being operable for applying the pulse and gradient switching sequence which produces a plurality of echoes.
  • the apparatus includes combining means for combining the echoes to obtain a combined data set and transforming means for transforming the combined data set into an NMR image of the object.
  • the sequence of stimulating signals which is applied preferably includes an excitation pulse such as a 90° radio frequency (rf) pulse, which rotates the resonant nuclei in the object through 90°.
  • an excitation pulse such as a 90° radio frequency (rf) pulse
  • rf radio frequency
  • This pulse is then followed by a series of refocusing pulses which will be 180° rf pulses, each of which rotates the nuclei through 180°.
  • the phase error introduced by the pulses is compensated if the excitation and refocusing pulses follow a specific sequence.
  • the specific pulse and gradient switching sequence may be described in terms of the axes along which the pulses and gradients are applied.
  • the axis along which the constant magnetic field is applied is generally designated the Z axis, and each plane perpendicular to that axis will have a corresponding X axis and Y axis.
  • the X and Y axes are riot stationary, however, but are rotating in a frame of reference known as the "rotating frame", which rotates at the nuclear spin resonant frequency. Therefore, the pulses and spin magnetizations which are applied along the the X and Y axes in the following discussion are actually applied in the rotating frame of reference.
  • the envelope of the given signal is at the nuclear spin resonant frequency, with its phase determined according to the axis along which it is applied.
  • a signal applied along the Y axis will accordingly be 90o out of phase with a signal applied along the X axis.
  • the compensating sequence of refocusing pulses according to the invention can be a series of 180° rf pulses applied in groups of 8 pulses along the positive and negative X and Y axes in the following order: +X, +Y, +X, +Y, -X, -Y, -X, -Y.
  • the pulses are applied alternately along the X and Y axes, and the first four are along the positive axes, while the second four are along the negative axes. If a sequence of stimulating signals like that above is applied, a series of groups of four echoes will be produced.
  • each group of four consecutive echoes two signals will have positive real parts and two will have negative real parts. Similarly, two signals will have positive imaginary parts and two will have negative imaginary parts. As a result, the echoes in each group may be combined to cancel any zero frequency or baseline components, which would otherwise produce an undesirable artifact.
  • each echo may be analyzed into an in phase magnetization component which has the same phase as the initial FID resulting from the excitation pulse and an out-of-phase component which is 90o out of phase from the initial FID.
  • the presence of the two components results from the application of phase encoding gradients, so that both components must be preserved in order to preserve the phase information. Therefore, the coherent addition of the echoes requires that both the in-phase components and the out-of-phase components may be added without cancellation to produce combined in-phase and out-of-phase components.
  • the process of coherent addition may be illustrated in terms of the pulse sequence described above.
  • the second of the echoes being added has an in-phase component which refocuses with inverted phase from that of the first echo, while its out-of-phase component refocuses with the same phase. Therefore, the first and second echoes may be coherently added after reversing the time axis of the second echo, multiplying it by -1 and taking its complex conjugate. As a result, each component of the second echo will be of the same sign as the corresponding component of the first, and the two echoes can be added together without cancelling.
  • alternate echoes may be time reversed and all of the even echoes may be added to form one data set while all of the odd echoes may be added to form another. In this case, the data set formed from the even echoes will contain information about spatial frequency from the positive phase encoding gradient, while the data set formed from the odd echoes will contain information about spatial frequency from the negative phase encoding gradient.
  • the echoes will be transformed, before or after coherent adding, to obtain a data set encoded according to spatial frequency.
  • This transform may be a Fourier transform or other similar transform, such as the FFT.
  • the terms "Fourier transform” and “transform” refer not only to the Fourier transform in its continuous and discrete forms, but also other similar transforms which depend on phase information, including digital approximations to the Fourier transform such as the FFT and the Hadamard transform. More generally, such a transform will be referred to herein as a "phase-dependent transform".
  • the data set encoded according to spatial frequency When the data set encoded according to spatial frequency has been completed, it may then be transformed to provide a spatial image of the object. This transformation may be in two or three dimensions, for example. Because the echoes are combined to form the data set, the resulting image has substantially better contrast than with the prior art techniques.
  • the method and apparatus of the present invention may reduce data acquisition time for NMR Fourier imaging. Rather than taking several separate FID measurements for signal averaging, as described above in relation to the prior art, the invention makes it possible to obtain several echoes from a single excitation. This is possible because the real spin-spin relaxation time, T 2, following an excitation is typically much longer than the effective relaxation time, T 2 *, of each echo. As a result, several echoes can be measured after each excitation and combined to obtain significantly greater sensitivity.
  • Fig. 1 is a cross sectional view of an arrangement of NMR gradient and rf coils around a subject.
  • Fig. 2A is a functional block diagram showing the basic operations of NMR Fourier imaging according to the invention.
  • Fig. 2B is a flow chart showing the basic steps of the pulse and gradient switching sequence of the invention.
  • Fig. 2C is a detailed flow chart showing one embodiment of coherent adding of echoes according to the invention.
  • Fig. 2D is a detailed flow chart showing an alternative embodiment of coherent adding of echoes according to the invention.
  • Fig. 3A is a timing diagram of a pulse and gradient switching sequence according to the invention.
  • Fig. 3B is a timing diagram of an alternative embodiment of the pulse and gradient switching sequence of the invention.
  • Fig. 4 is a graph which compares the pulse error compensation resulting from the pulse and gradient switching sequence of Fig. 3A with the pulse error which would result from the use of a prior art pulse sequence.
  • Fig. 5A is a graphical illustration of the real and imaginary parts of an odd echo.
  • Fig. 5B is a graphical illustration like that of
  • Fig. 5C is a graphical illustration of the time reversal of the waveform of Fig. 5B.
  • Fig. 5D is a graphical illustration of the complex conjugate of the waveform of Fig. 5C.
  • Fig. 6 is a graphical illustration of a two-dimensional data set derived from the coherent addition of echoes.
  • Fig. 7 is a simplified representation of an image which would result from the transformation of the data set shown in Fig. 6.
  • Fig. 8 is a block diagram of the apparatus of the invention.
  • Fig. 9 is a graph showing the improvement of the relative signal-to-noise ratio as the number of echoes sampled increases.
  • Fig. 10 is a sequence of graphs showing the phases of the components of a series of echoes.
  • Fig. 11 is a sequence of graphs showing in greater detail the production of the first two echoes.
  • the invention provides a method and apparatus for generating an NMR image of an object.
  • the general features of the invention can be understood from Figs. 1 and 2A.
  • Fig. 1 shows gradient and radio frequency (rf) coils mounted in relation to a human body for imaging.
  • Magnetic coils 10 are oriented in relation to X, Y and Z laboratory axes, which are stationary.
  • Static coil 12 is a solenoidal magnet which will conduct a current to generate a static, uniform magnetic field B o along the Z axis, surrounding body 18.
  • Gradient coils 14, 15 and 16 are arranged to provide a gradient field combining components
  • rf coil 17 is tuned to the resonant frequency of the substance being imaged. Therefore, rf coil 17 may operate both to provide rf pulses and to detect rf signals from body 18. These signals may then be processed to provide information about body 18, such as an image.
  • Fig. 2A is a functional block diagram showing the basic operations of NMR Fourier imaging 20 according to the invention. Block 30 illustrates that the first operation is to apply the pulse and gradient switching sequence. The gradients are applied using gradient coils 14, 15, 16 as shown in Fig. 1, while the pulses are applied using rf coil 17. As shown in block 40, if the sequence of pulses and gradients is appropriate, rf coil 17 will receive echoes containing both phase and amplitude information.
  • the echoes When the echoes have been received as indicated in block 40, they may be combined to form a combined data set, as shown in block 50. This process of combining the echoes must similarly be performed in a manner which preserves the phase information, so that the data set can then be transformed into an image, as shown in block 60.
  • the pulse and gradient switching sequence applied in block 30 is an important factor in determining whether the subsequent steps of the invention will lead to a high-quality image.
  • the invention provides a pulse and gradient switching sequence which is effective to preserve the phase information and to provide sequences of echoes which can be readily combined without loss of phase information.
  • the pulse and gradient switching sequence of the invention provides a series of NMR echoes which can be combined into a single data set without loss of phase information.
  • Fig. 2B is a flow chart showing the basic steps of a pulse and gradient switching sequence according to the invention, while Figs. 3A and 3B are timing diagrams of two different pulse and gradient switching sequences according to the invention.
  • the pulse and gradient switching sequence must stimulate a resonance whose real decay time is longer than its effective decay time.
  • a spin-spin relaxation time, T 2 of each atom may be substantially longer than the effective spin-spin relaxation time, T 2 *, observed for a part of the object containing many atoms.
  • T 2 * effective spin-spin relaxation time
  • the pulse and gradient switching sequence 30 of the invention begins with a 90° rf excitation pulse aiong the Y axis, shown in block 32.
  • This excitation pulse begins the T 2 decay of each atom's resonance. It also begins an initial T 2 * decay of the composite of the atoms. This can be seen in Fig. 3A, in which the excitation pulse 72 is followed by an initial T 2 * decay 74.
  • the first of the echoes is produced by the first refocusing pulse, shown in block 34a in Fig. 2B.
  • the mechanism by which the refocusing pulses produce echoes will be discussed in greater detail below in relation to Figs. 10 and 11.
  • the excitation pulse is a 90° pulse along the positive Y axis
  • the first refocusing pulse of pulse and gradient switching sequence 30 will be a 180° rf pulse along the X axis. This pulse serves to reflect the spin magnetization vectors around the X axis, so that as the T 2 decay continues, the magnetization is refocused to form a spin echo decaying according to T 2 *.
  • the excitation pulse is a 90° pulse along the positive Y axis
  • the first refocusing pulse of pulse and gradient switching sequence 30 will be a 180° rf pulse along the X axis.
  • This pulse serves to reflect the spin magnetization vectors around the X axis, so that as the T 2 decay continues, the magnetization is refocused
  • the first refocusing pulse 76 will result in an echo 78 centered about the point at which the axes of the atoms are aligned approximately as they were after the excitation pulse 72.
  • echo 78 rises to its peak along a wave form which resembles a mirror image of the initial T 2 * decay 74. Echo 78 then decays with a time constant of T 2 *, again resembling the wave form of initial T 2 * decay 74.
  • a second refocusing pulse a 180° rf pulse along the Y axis is applied, as shown in block 34b of Fig. 2B.
  • the effect of this pulse is to reflect the spin axes of the atoms about the Y axis, resulting in another echo.
  • the second refocusing pulse 76' results in a second echo 78', closely resembling the first echo 78.
  • echoes can be formed by reversing the gradient along the Y axis G y .
  • the sign of the gradients may be alternately positive and negative. This results in a series of echoes which can be coherently added, but which suffer from progressive dephasing as a result of phase error. Therefore, this technique is not as desirable as the use of the pulse and gradient switching sequence shown in Fig. 3A.
  • Pulse and gradient switching sequence 30 continues with a third, refocusing pulse, as shown in block 34c in Fig. 2B.
  • This third refocusing pulse is a 180o rf pulse along the X axis.
  • a fourth refocusing pulse is applied as shown in block 34d in the form of a 180o rf pulse along the Y axis.
  • the result of the first through fourth refocusing pulses 34a-34d will be to return the spin axes of the atoms to the position prior to the first refocusing pulse 76.
  • the phase error introduced by each of the refocusing pulses will be compensated.
  • the pulse and gradient switching sequence 30 of the invention also includes a series of refocusing pulses along the negative X and Y axes, as shown in Fig. 2B.
  • the fifth refocusing pulse, shown in block 36a is a 180° rf pulse along the negative X axis
  • the sixth refocusing pulse, shown in block 36b is a 180o rf pulse along the Y axis.
  • the seventh and eighth refocusing pulses shown in blocks 36c and 36d are also 180o rf pulses along the negative X and Y axes. Therefore, the fifth through eighth refocusing pulses compensate pulse phase error in the same manner as the first through fourth refocusing pulses shown in blocks 34a-34d.
  • the sequence of refocusing pulses in blocks 34a-34d and 36a-36d can be repeated in the order given to obtain additional echoes which are compensated.
  • the number of refocusing pulses possible will depend on the ratio between T 2 and T 2 *, discussed in greater detail below. If that ratio is quite large, a large number of echoes may be obtained by repeating the sequence of eight refocusing pulses several times. It is not necessary, however, to repeat the entire sequence of eight pulses to obtain the compensation of phase error, as will be seen from Fig. 4A, discussed below.
  • Figs. 4A and 4B compares the phase error after each pulse of the pulse and switching sequence 30 (the XYXY sequence) with the error after each pulse of the Carr-Purcell sequence (CP) and the Carr-Pur- cell-Meiboom-Gill sequence (CPMG) used in some types of NMR measurement to obtain spin echoes.
  • CP Carr-Purcell sequence
  • CPMG Carr-Pur- cell-Meiboom-Gill sequence
  • S. Meiboom, and D. Gill, 29 Rev. Sci. Instruments 688 (1958) discuss these sequences.
  • Fig. 4A shows the variation in amplitude of the relative magnetization of the in-phase and out-of-phase components of the echoes resulting from the pulse and gradient switching sequence 30 of the invention.
  • Fig, 4B shows the variation in amplitude of the in-phase and out-of-phase components of the echoes resulting from the CP and CPMG pulse sequences.
  • each refocusing pulse rather than being a full 180o pulse, is an effective pulse of 170o, and therefore introduces an error of 10o per pulse. This error may result from inaccurate setting or other causes.
  • both the in-phase and out-of-phase components of the echoes are fully compensated for the error after each group of eight spin echoes.
  • both components have values close to the full relative magnetization even for those echoes which are not fully compensated, with the maximum loss of signal less than 8%.
  • the in-phase components of the echoes resulting from the CP and CPMG pulse sequences are compensated, but the out-of- phase components deteriorate rapidly.
  • the in-phase CP and CPMG components are very well compensated, with a loss of 1.5% on alternate refocusing pulses.
  • the out-of-phase CP and CPMG components show no substantial compensation, so that the amplitude varies widely. After nine refocusing pulses, each in error by 10°, the amplitude goes to zero, indicating that the full 90° of error is uncompensated. As a result, the phase information is lost with the CP and CPMG pulse sequences.
  • the refocusing pulses of the invention are applied along X and Y axes in the rotating frame of reference.
  • gradient fields along the X, Y and Z axes in the laboratory frame of reference may also be applied for spatial encoding of the NMR data.
  • Figs. 3A and 3B show in greater detail the timing of the pulses discussed above.
  • gradients 82 and 84 are applied after the excitation pulse 72.
  • Gradient 82 may include both G x and G z components. These components of gradient 82 are applied only during the evolution period between initial excitation pulse 72 and the first refocusing pulse 76.
  • the spin axes of the atoms begin in a plane perpendicular to the Y axis and move toward the Y axes at a rate determined by gradient 84 which has a component G y .
  • Gradient 82 performs a phase encoding function, which encodes the phase of the magnetization in the X-Z plane according to the spatial location along the X and Z axes in the laboratory frame of reference. The amount of phase encoding depends on the strength of the components of the field gradient 82 and the location of the spins in the
  • the G x and G z components of gradient 82 may be applied successively.
  • the G y component of gradient 84 is necessary because refocusing gradients 86 and 86' also include G y components.
  • gradient 84 and 86 must satisfy the condition expressed in the following equation:
  • t represents the time from excitation pulse 72
  • t 1 represents the time from the peak of the first echo 78
  • t 1 represents the time from the peak of the first echo 78
  • t 2 represents the time between the excitation pulse 72 and the first refocusing pulse 76.
  • ⁇ is chosen to provide an optimal evolution, and the time between refocusing pulses will be 2 .
  • the first half of each refocusing gradient 86, 86' is a phase encoding gradient over time, while the second half serves to re-encode the phase of the magnetization along the Y axis. Because of this use of G y for two purposes, gradient 84 is necessary for encoding before the echoes begin.
  • gradient 88 of Fig. 3B contains only a G x component.
  • the pulse and gradient switching sequence of Fig. 3B does not result in phase encoding in the Z direction.
  • This sequence can be used, however, for two-dimensional imaging, in which a plane transverse to the Z axis is selected by a selective pulse 87 and appropriate refocusing gradients 89 and 89' are applied, all being G z components.
  • the sequence shown in Fig. 3B is the same as that of Fig. 3A.
  • a single excitation pulse 72 is followed by refocusing pulses, including first refocusing pulse 76 and second refocusing pulse 76'.
  • refocusing pulses including first refocusing pulse 76 and second refocusing pulse 76'.
  • a series of observations must be made, each beginning with an excitation pulse 72 followed by gradient 82 or gradient 88.
  • the phase encoding which may be measured in degrees, depends on the strength of these gradients. Therefore, a series of measurements may be taken using gradients of different strengths to obtain data sets with different amounts of phase encoding. These data sets may then be ordered according to phase encoding to provide a combined data set which is then transformed.
  • the pulse and gradient switching sequence 30 of the invention provides a series of echoes having specific characteristics.
  • the application of a series of pulse and gradient switching sequences 30 with different degrees of phase encoding will produce a set of echo trains, each of which may be combined to form one or more data sets according to the invention.
  • a series of pulse and gradient switching sequences 30 with different degrees of phase encoding will produce a set of echo trains, each of which may be combined to form one or more data sets according to the invention.
  • Equation (2) the signal is expressed as S, M 0 represents the equilibrium magnetization at the point x 0 , y 0 , z0 , and w is equal to the product of the gyromagnetic ratio ⁇ with the coordinate z 0 and the gradient G z .
  • ⁇ is the phase encoding term, given by:
  • each signal may be expressed as the sum of a real term and a complex term.
  • Each of these terms will be dependent on the phase encoding term ⁇ , resulting from the successive applications of G x and G z , as they affect the location x 0 ,
  • the resulting spin packet at the location x 0 , y 0 , z 0 will be the sum of all spin magnetization arising in a voxel at that position, and the phase of this spin packet may vary over a full 360° or 2 ⁇ radians.
  • Signals S A and S B are orthogonal, and together are equal to the entire signal S.
  • S A and S B may be obtained from a phase sensitive detector. Therefore, the signals obtained at the first echo for each of these components will be:
  • Fig. 10 shows schematically the magnetization resulting from some pulses of pulse and gradient switching sequence 30.
  • excitation pulse 72 a 90° rf pulse along the Y axis in the rotating frame of reference, causes the magnetization to rotate through 90° from the Z axis to the X axis, resulting in initial magnetization 110.
  • initial magnetization 110 evolves into a form having in-phase component A and out-of-phase component B.
  • first refocusing pulse 76a a 180° rf pulse along the X axis
  • first echo magnetizations 112 and 122 result along the X and Y axes, respectively.
  • first echo magnetization 112 of the in-phase component A is illustrated schematically in the top line of Fig. 11.
  • excitation pulse 72 rotates the magnetization through 90° to form initial magnetization 110.
  • magnetization 110 will dephase, resulting in initial T 2 * decay 74, shown in Fig. 3A.
  • This dephasing is illustrated in Fig. 11 by two representative spin vectors 110a and 110b which, even though still excited, tend to cancel each other due to their movements in different directions.
  • refocusing pulse 76a is applied, however, vectors 110a and 110b are reflected about the X axis, creating reflected vectors 112a and 112b. Reflected vectors 112a and 112b then refocus by converging to form first echo magnetization 112.
  • third refocusing pulse 76c a 180o rf pulse along the X axis, will invert the phase of only the out-of-phase component, as third echo magnetization 126, leaving the phase of third echo magnetization 116 the same for the in-phase component.
  • Fourth refocusing pulse 76d a 180° rf pulse along the Y axis, will invert the in-phase fourth echo magnetization 118 but not the out-of-phase magnetization 128.
  • the components A and B of the fourth echo will have the same phase relation to the components of the third echo as the components of the second echo have to the first echo.
  • the echo from each even-numbered refocusing pulse of pulse and gradient switching sequence 30 will have this same relation to the preceding odd-numbered echo, so that we may generally treat equations 4A and 4B as describing the odd echoes and equations 5A and 5B as describing the even echoes.
  • each echo received as a result of the pulse and gradient switching sequence 30 contains both amplitude and phase information which can be represented by the real and imaginary components of a complex function.
  • Figs. 2C and 2D are flow charts showing two ways in which these echoes may be coherently added according to the invention. In both cases, the coherent addition results in a data set in which both phase and amplitude information are present.
  • the process begins with block 42, in which the echoes are received and converted into stored digital signals. This may be done, for example, by a phase-sensitive detector followed by an analog-to-digital converter.
  • the result will preferably be a group of measurements for the in-phase and out-of-phase components similar to those in equations 4 and 5, above.
  • the out-of-phase components represented in equations 4B and 5B
  • this can be done by time reversing the alternate echoes, as shown in block 52. If we treat the second echo as the echo to be reversed in time, we may rewrite equations 5A and 5B by replacing t 2 with -t 2 as follows:
  • the next step, shown in block 54, is to take the complex conjugate of the alternate echoes, which yields:
  • equations 7A and 7B As can be seen by comparing equations 7A and 7B with equations 4A and 4B, the signs of the terms are now the same for the first echo and the second echo, and the two data sets can be added together, term by term. Furthermore, subsequent echoes can be treated in the same manner such that all odd numbered echoes are treated alike and all even numbered echoes are also treated alike. Therefore, the final step of this embodiment is to add the echoes as shown in blcck 56 of Fig. 2C. Since the time reversal of block 52 and the complex conjugate of block 54 do not affect the sign of the real components, the addition of the echoes in block 56 also results in a cancellation of any existing zero frequency or baseline components.
  • the real parts of the complex data set will be summed with the real parts of the complex data set will be summed with the signs (+, -, -, +) for each group of four echoes. Since the complex conjugate operation of block 54 changes the sign of imaginary zero frequency components, they will be summed with the signs (+, +, -, -), which will also result in cancellation of the baseline offsets. Therefore, the use of the pulse and gradient switching sequence 30 with coherent addition 50 results in cancellation of baseline offsets.
  • Figs. 5A-5D illustrate graphically the process of coherent addition according to Fig. 2C.
  • the in-phase signal S A is shown in solid line, while the out-of-phase signal S B is shown in dashed line.
  • Fig. 5A shows real components 42a and imaginary components 42b of an odd echo, each with an arbitrary illustrative waveshape.
  • Fig. 5B shows real components 42c and imaginary components 42d of an even echo, which will bear a characteristic relation to the components 42a and 42b of the odd echo.
  • Fig. 5C shows the even echo of Fig.
  • Fig. 5B shows the even echo after taking the complex conjugate of the components of Fig. 5C, so that real components 54a are the same as components 52a, while imaginary components 54b are opposite in sign from components 52b.
  • the real components 42a and 54a are similar in shape while imaginary components 42b and 54b are also similar, illustrating graphically how the odd and even echoes can be added without cancellation.
  • Fig. 3A shows gradient 82 as a combination of positive amplitudes, for illustrative purposes.
  • the second echo is also modified and becomes:
  • Equation 9A and 9B Comparing equations 9A and 9B with equations 4A and 4B, it is apparent that the second echo obtained from the negative application of the spatial encoding gradient is identical, except for relaxation effects, to the first echo from the positive application of the same spatial encoding gradient. Therefore, these two data sets may be added together without any additional data processing. Similarly, comparing equations 8A and 8B with equations 5A and 5B, it is apparent that the first echo from the negative spatial encoding gradient may be added to the second echo from the positive encoding gradient without additional data processing.
  • Fig. 2D shows an alternative method of coherently adding the echoes to form data sets which follows from the above discussion. In this method, the echoes are received and converted to stored digital signals in block 42, in the same manner as in Fig.
  • the first data set from each phase encoding gradient will be added to the second data set from the phase encoding gradient of opposite sign, resulting in two combined data sets.
  • a Fourier transformation is performed to convert from the time domain to the frequency domain.
  • the data sets can be ordered according to the phase encoding into a combined data set.
  • An example of such a combined data set 60 is shown graphically in Fig. 6.
  • the Fourier transformation of this combined data set will then produce a spatial image of the object under investigation.
  • Fig. 7 shows a simplified representation of an image 62 which would result from taking the Fourier transform of the data set 60 shown in Fig. 6.
  • NMR Fourier Imaging Apparatus Fig. 8 shows a block diagram of the apparatus of the invention which may be used for performing NMR Fourier imaging.
  • NMR Fourier imaging apparatus 100 includes control and processing circuitry 90, which may typically be a programmed digital computer.
  • control and processing circuitry 90 will be connected to gradient coil control circuitry 102a and rf pulse control circuitry 102b.
  • computer 90 will be connected to receiving circuitry 102c for receiving the NMR signals and converting them into digital form.
  • computer 90 is represented as including a central processing unit (CPU) 92, program memory 94, and a data memory 96.
  • the operation of the CPU 92 includes sending and receiving appropriate signals from the gradient coil control circuitry 102a, the rf pulse control circuitry 102b and the receiving circuitry 102c.
  • CPU 92 is connected to an appropriate operator interface 104, which may be a standard operator terminal.
  • CPU 92 is connected to a display device 106, which may be a CRT display or any other appropriate display having the necessary degree of resolution.
  • CPU 92 will be initiated by appropriate signals from the operator interface 104. In response to those signals, CPU 92 will initiate the pulse and gradient switching sequence 30 in response to the switching sequence program 94a stored in program memory 94. Execution of this program will cause CPU 92 to send signals to gradient coil control circuitry 102a and rf pulse control circuitry 102b to produce the pulse and gradient switching sequence 30. As a result, echoes will be received by receiving circuitry 102c and changed into digital form, as described above. As a part of the switching sequence program 94a, CPU 92 also operates to store the digitized echoes in the echo data memory 96a.
  • CPU 92 also executes coherent addition program 94b in order to combine each set of echoes into a respective data set stored in combined data set memory 96b.
  • the echo data will be Fourier transformed before or after the coherent addition, and this Fourier transformation should be performed as a part of coherent addition program 94b.
  • CPU 92 executes transform algorithm 94c in order to obtain the basic image data which is stored in display data memory 96c.
  • the display data may then be read out by CPU 92 to the display device 106 for display to the operator.
  • the switching sequence program 94a and coherent addition program 94b will perform the steps described above in order to preserve the phase information necessary to perform Fourier imaging.
  • these programs will preferably conform to the additional design constraints set forth below.
  • T 2 * used above to represent the apparent relaxation time, may be more narrowly defined as the relaxation time which takes into account the additional contribution from local field inhomogeneity over the resolution element or voxel in question.
  • T 2 ** can then be defined as the relaxation time which takes into account the field inhomogeneity over the whole object within the imaging region.
  • t s should be as short as possible, and should therefore have a value of twice T 2 **. This will result in an image which is relatively free of spatial distortions due to field inhomogeneity.
  • T 2 * places a lower limit on the strength of the field gradient necessary to produce an acceptable image.
  • the image will, however, be spatially distorted due to the effect of field inhomogeneity. This distortion may be acceptable for the particular imaging requirements or, alternatively, the distortion may be corrected by suitable software.
  • the S/N ratio may be improved by reducing the band width over which the NMR signal is received, because the noise power at each point is proportional to the square root of the frequency band at that point.
  • the frequency band is the reciprocal of the sampling time t s .
  • the signal S(M) which can be obtained by summing echoes, keeping all other parameters constant, is given by:
  • Equation 13 we can obtain an equation for the relative improvement R(M), which is the ratio between the S/N ratio with M echoes to the S/N ratio from the first echo alone.
  • Equation 14 is plotted in Fig. 9 for several values of T 2 /T 2 **, illustrating the relative S/N ratio resulting from the addition of various numbers of echoes for each value of T 2/T 2 **. If T 2 /T 2 ** > 1, some gain in sensitivity is possible by adding echoes.
  • the operator may vary the value of T 2 /T 2 ** to some extent by the choice of resolution and by setting the acceptable degree of spatial distortion. This ratio becomes particularly important at higher field strengths, since magnets for imaging applications are pres ently specified with the same order of inhomogeneity regardless of the maximum field strength.
  • the ratio of T 2 /T 2 ** is much larger at higher fields than at lower fields, and the value of R(M) in equation 14 is increased.
  • T 2 /T 2 ** 3
  • the fourth echo has an intensity of 37% of the first echo and a sensitivity improvement of 30% is achieved by sampling four echoes.
  • T 2 /T 2 ** 30
  • a sensitivity improvement of 355% is achievable by sampling 38 echoes.

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

Des images de Fourier basées sur des échos NMR (résonance magnétique nucléaire) multiples peuvent être obtenues en utilisant le procédé et l'appareil ci-décrit. Une séquence d'impulsions et de gradients sont appliqués à l'objet à prendre en image (30). Les impulsions commencent avec une impulsion d'excitation (72) qui est suivie d'une série d'impulsions de refocalisation (76, 76'). De manière à compenser l'erreur de phase dans les impulsions de refocalisation, celles-ci sont appliquées le long des axes alternatifs X et Y dans le cadre rotatif de référence. Les échos (78, 78') sont ensuite additionnés de manière cohérente pour conserver les informations de phase (56, 56'). Des échos alternés peuvent être inversés (52) dans le temps et leur conjugué complexe peut être pris (54) pour permettre une addition cohérente (56). Alternativement, les échos pairs (54') peuvent être ajoutés séparément par rapport aux échos impairs (56') pour produire des ensembles de données séparés. La séquence de commutation d'impulsions de gradients (30) fournit des inversions de signes de sorte que toute composante de ligne de base est annulée par l'addition cohérente. Avant ou après une addition cohérente, les échos sont transformés, et après formation d'un ensemble de données combinées à partir d'une série de séquences d'échos (50), l'ensemble de données combinées est de nouveau transformé pour produire une image de l'objet (60).
PCT/US1985/000950 1984-06-01 1985-05-21 Prise d'image de fourier nmr a partir d'echos multiples WO1985005693A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US61628384A 1984-06-01 1984-06-01
US616,283 1984-06-01

Publications (1)

Publication Number Publication Date
WO1985005693A1 true WO1985005693A1 (fr) 1985-12-19

Family

ID=24468776

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1985/000950 WO1985005693A1 (fr) 1984-06-01 1985-05-21 Prise d'image de fourier nmr a partir d'echos multiples

Country Status (2)

Country Link
EP (1) EP0182873A1 (fr)
WO (1) WO1985005693A1 (fr)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4684890A (en) * 1984-12-12 1987-08-04 Commissariat A L'energie Atomique Process for the generation and processing of signals for obtaining by nuclear magnetic resonance a distortion-free image from a heterogeneous polarization field
US4689568A (en) * 1985-03-11 1987-08-25 Hitachi, Ltd. NMR chemical shift imaging method
DE3710748A1 (de) * 1986-03-31 1987-10-08 Toshiba Kawasaki Kk Magnetresonanz-abbildungsanordnung
EP0250050A1 (fr) * 1986-06-18 1987-12-23 Koninklijke Philips Electronics N.V. Procédé pour reconstruire la distribution de la magnétisation nucléaire par la mesure partielle de la résonance magnétique
EP0296834A2 (fr) * 1987-06-24 1988-12-28 Picker International, Inc. Procédé et appareil pour l'imagerie par résonance magnétique
EP0339979A2 (fr) * 1988-04-27 1989-11-02 Btg International Limited Système d'imagerie par RMN
EP0383631A2 (fr) * 1989-02-17 1990-08-22 Kabushiki Kaisha Toshiba Système d'imagerie par résonance magnétique
WO1994004937A1 (fr) * 1992-08-20 1994-03-03 British Technology Group Limited Procede et dispositif d'analyse par resonance magnetique nucleaire
EP0610012A1 (fr) * 1993-02-01 1994-08-10 Panacea Medical Laboratories Système à positionnement commandable à distance pour l'imagerie par résonance magnétique

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115951281B (zh) * 2022-12-08 2023-10-20 无锡鸣石峻致医疗科技有限公司 磁共振梯度延迟和梯度切换率同时检测方法、装置和介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4297637A (en) * 1978-07-20 1981-10-27 The Regents Of The University Of California Method and apparatus for mapping lines of nuclear density within an object using nuclear magnetic resonance
US4390840A (en) * 1979-12-20 1983-06-28 Siemens Aktiengesellschaft Zeugmatography process
US4431968A (en) * 1982-04-05 1984-02-14 General Electric Company Method of three-dimensional NMR imaging using selective excitation

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4297637A (en) * 1978-07-20 1981-10-27 The Regents Of The University Of California Method and apparatus for mapping lines of nuclear density within an object using nuclear magnetic resonance
US4390840A (en) * 1979-12-20 1983-06-28 Siemens Aktiengesellschaft Zeugmatography process
US4431968A (en) * 1982-04-05 1984-02-14 General Electric Company Method of three-dimensional NMR imaging using selective excitation

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4684890A (en) * 1984-12-12 1987-08-04 Commissariat A L'energie Atomique Process for the generation and processing of signals for obtaining by nuclear magnetic resonance a distortion-free image from a heterogeneous polarization field
US4689568A (en) * 1985-03-11 1987-08-25 Hitachi, Ltd. NMR chemical shift imaging method
DE3710748A1 (de) * 1986-03-31 1987-10-08 Toshiba Kawasaki Kk Magnetresonanz-abbildungsanordnung
EP0250050A1 (fr) * 1986-06-18 1987-12-23 Koninklijke Philips Electronics N.V. Procédé pour reconstruire la distribution de la magnétisation nucléaire par la mesure partielle de la résonance magnétique
EP0296834A2 (fr) * 1987-06-24 1988-12-28 Picker International, Inc. Procédé et appareil pour l'imagerie par résonance magnétique
EP0296834A3 (en) * 1987-06-24 1990-11-28 Picker International, Inc. Magnetic resonance imaging methods and apparatus
US5051698A (en) * 1988-04-27 1991-09-24 National Research Development Corporation NMR imaging systems
EP0339979A2 (fr) * 1988-04-27 1989-11-02 Btg International Limited Système d'imagerie par RMN
EP0339979A3 (fr) * 1988-04-27 1990-12-19 Btg International Limited Système d'imagerie par RMN
EP0383631A2 (fr) * 1989-02-17 1990-08-22 Kabushiki Kaisha Toshiba Système d'imagerie par résonance magnétique
EP0383631A3 (fr) * 1989-02-17 1991-05-08 Kabushiki Kaisha Toshiba Système d'imagerie par résonance magnétique
US5055789A (en) * 1989-02-17 1991-10-08 Kabushiki Kaisha Toshiba Magnetic resonance imaging system
WO1994004937A1 (fr) * 1992-08-20 1994-03-03 British Technology Group Limited Procede et dispositif d'analyse par resonance magnetique nucleaire
EP0610012A1 (fr) * 1993-02-01 1994-08-10 Panacea Medical Laboratories Système à positionnement commandable à distance pour l'imagerie par résonance magnétique

Also Published As

Publication number Publication date
EP0182873A1 (fr) 1986-06-04

Similar Documents

Publication Publication Date Title
US4689567A (en) NMR Fourier imaging from multiple echoes
US5942897A (en) Magnetic resonance imaging apparatus
US4748410A (en) Rapid NMR imaging system
US7068031B2 (en) MR imaging method and MRI system
EP2233940B1 (fr) Excitation et acquisition simultanée en résonance magnétique
US4691162A (en) Method and device for NMR spectroscopy
JP3734086B2 (ja) 核磁気共鳴イメージング装置
JP3529446B2 (ja) Ｅｐｉ及びｇｒａｓｅ ｍｒｉにおける読み出し傾斜磁界極性の補正方法
US4993075A (en) Image reconstruction method in NMR imaging
US5652516A (en) Spectroscopic magnetic resonance imaging using spiral trajectories
JPH0224135B2 (fr)
US4628262A (en) Multiple echo chemical shift imaging
EP0145276B1 (fr) Représentation du quotient complexe en résonance magnétique nucléaire
US5025216A (en) Magnetic resonance imaging of short T2 species
US4745364A (en) Method and apparatus for rapid NMR imaging
US4985677A (en) Magnetic resonance imaging and spectroscopy using an excitation pulse for multiple-dimensional selectivity
EP0414318B1 (fr) Méthode multimode rapide d'imagerie par résonance magnétique
US4868501A (en) Method and means for magnetic resonance spin-echo imaging using an adiabatic three pi pulse sequence
US4654591A (en) NMR flow imaging using bi-phasic excitation field gradients
US4706027A (en) Method for correcting phase errors in magnetic resonance imaging data
WO1985005693A1 (fr) Prise d'image de fourier nmr a partir d'echos multiples
US5570020A (en) Mr method and device for carrying out such a method
CN110730912B (zh) 并行多切片mr成像
US4697149A (en) NMR flow imaging using a composite excitation field and magnetic field gradient sequence
US11543482B2 (en) Magnetic resonance imaging using motion-compensated image reconstruction

Legal Events

Date Code Title Description
AK Designated states

Designated state(s): JP

AL Designated countries for regional patents

Designated state(s): AT BE CH DE FR GB IT LU NL SE