WO2000037956A1

WO2000037956A1 - Enhanced t2*-contrast in half-fourier magnetic resonance imaging

Info

Publication number: WO2000037956A1
Application number: PCT/EP1999/009819
Authority: WO
Inventors: Johan S. Van Den Brink
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 1998-12-22
Filing date: 1999-12-13
Publication date: 2000-06-29

Abstract

The invention relates to an MR method for imaging an object. During measurement of MR signals by application of gradients, a measuring plane is sampled along a trajectory which is asymmetrically situated in the k-space and contains a center of the k-space. In order to reduce the acquisition time while retaining a long echo time, the trajectory in the k-space is sampled in such a manner that a distance between a starting point and the center of the k-space is greater than a distance between an end point and the center of the k-space.

Description

ENHANCED T2*-CONTRAST IN HALF-FOURIER MAGNETIC RESONANCE IMAGING

The invention relates to a method for imaging an object by means of magnetic resonance (MR), which method involves an imaging pulse sequence which includes the following steps: generating an excitation RF pulse for the excitation of spins in the object, - measuπng position-dependent MR signals duπng an acquisition by application of at least one read-out gradient along a trajectory in a measuπng plane a k-space. the trajectory containing a center of the k-space and the measuπng plane being asymmetπcally situated relative to the center of the k-space.

The invention also relates to a device for carrying out such a method. In the context of the present application the k-space is to be understood to mean a spatial frequency domain in which a trajectory is completed by application of the gradients duπng the measurement of MR signals. The trajectory is determined by the time integral of the gradients applied in the interval from the excitation RF pulse until the instant at which the MR signal is measured. The measured values of the MR signals correspond to the inverse Fouπer transformed values of an image of the object In the context of the present application gradients are to be understood to mean temporary magnetic fields which are superposed on the steady magnetic field and cause magnetic field gradients in three orthogonal directions

A method and a device of the kind set forth are known from United States patent 4,767,991. A method of this kind can be used inter aha for the formation of in vivo functional MR images, venographic MR images and thermographic MR images The Blood Oxygen Level Dependent (BOLD) contrast can be used in functional MR images, for example in order to distinguish vaπous functional areas of brain tissue of a body to be examined According to the known method MR signals are measured along the trajectory in the k-space, so that the measuπng plane in the k-space is uniformly sampled In order to reduce the acquisition time duπng which the MR signals are measured, the measuπng plane is asymmetπcally situated in the k-space. For a better signal-to-noise ratio of the measured MR signals, the applied phase encoding and read-out gradients are chosen to be such that the trajectory passes through the center of the k-space at the beginning of the acquisition. The measured MR signals are subsequently disposed in a part of a symmetπcal matπx, corresponding to the trajectory, in order to reconstruct an MR image, for example by means of two-dimensional Fouπer transformation. For example, at least 50% (but more m practice) of the symmetπcal matπx is filled in this manner. In order to reconstruct an MR image which is suitable for diagnostic purposes, complex conjugate values of the measured MR signals are inserted in a remainder of the symmetπcal matπx in order to make the matπx symmetπcal. It is a drawback of the known method that the MR images exhibit a low T_' contrast between different parts of the object to be imaged. The T₂ ^* contrast is of importance for functional MR images and venographic MR images. The T_^* contrast is also of importance for the Blood Oxygen Level Dependent (BOLD) contrast in functional MR images.

It is an object of the invention to provide a method in which the T_^* contrast in the MR image is enhanced. To achieve this, the method according to the invention is characteπzed in that a distance between a starting point of the trajectory and the center is greater than a distance between an end point of the trajectory and the center of the k-space. As a result of this choice of the starting point and the end point of the trajectory, in compaπson with the echo time of the known method the echo time is now shifted, viewed in time, towards the end of the acquisition. The echo time is defined by the interval which elapses between the instant of application of the excitation RF pulse and the instant at which the trajectory passes the center in the k-space. Because for different tissues the differences in respect of T_^* decay increase as a function of the length of the echo time, the T contrast in the MR image will increase It is a further advantage that a phase change caused by proton chemical shift can be more accurately measured when the echo time T_E IS longer This effect is used so as to form thermographic MR images.

A special version of the method according to the invention is characteπzed in that the k-space contains two orthogonal axes, and that one of the axes of the k-space corresponds to a phase encoding gradient. The number of measurements by means of a varying phase encoding gradient is thus limited.

The invention is also advantageously employed in conjunction with radially scanning of k-space. Notably, scanning is performed along a plurality of lines through the oπgmal (k_x=0, k_y=0) of the (k_n.k_y) plane, where individual lines have respective directions in k-space. According to the invention, MR signals are measured along stretches of those lines which extend over different distances into opposite quadrants in k-space. A further version of the method according to the invention is characterized in that the k-space contains two orthogonal axes, and that one of the axes of the k-space corresponds to the read-out gradient. The number of measurements with one value of the phase encoding gradient is thus limited. A further version of the method according to the invention is characterized in that in the imaging pulse sequence at least one refocusing RF pulse is generated subsequent to the excitation RF pulse. The use of one or more refocusing RF pulses counteracts disturbances in the MR signals due to inhomogeneities in the steady magnetic field.

A further version of the method according to the invention is characterized in that the imaging pulse sequence is repeated, a selection gradient being applied upon generation of the excitation RF pulse and a first additional selection gradient being applied after the selection gradient in order to avoid a net dephasing of the spins excited during a previous imaging pulse sequence, after which a second additional selection gradient is applied, after the measurement of MR signals, in the imaging pulse sequence in order to rephase the spins excited during the imaging pulse sequence. A selection gradient is to be understood to mean one of the gradients applied simultaneously with the generation of the excitation RF pulse in order to select a region within the object. A method of this kind is known from the article "A functional MRI technique combining principles of echo-shifting with a train of observations (PRESTO)", by G. Liu et al., published in "Magnetic Resonance in Medicine", 30, pp. 764- 768 (1993). As a result of the application of the additional selection gradients, an echo of the spins excited in the imaging pulse sequence is shifted to a next imaging pulse sequence. In comparison with a conventional gradient echo method this method offers a reduced acquisition time without reducing the echo time.

The invention also relates to a device for carrying out such a method. The device according to the invention is characterized in that the control unit is also arranged in such a manner that a distance between a starting point of the trajectory in the k-space and the center is greater than a distance between an end point of the trajectory and the center.

These and other aspects of the invention are apparent from and will be elucidated, by way of example, with reference to the embodiment described hereinafter and shown in drawing.

In the drawing:

Fig. 1 shows a magnetic resonance device,

Fig. 2 shows an EPI imaging pulse sequence,

Fig. 3 shows a first measuring plane and a first trajectory in the k-space, Fig. 4 shows a PRESTO imaging pulse sequence, and Fig 5 shows a second measuπng plane and a second trajectory in the k-space Fig. 1 shows a magnetic resonance imaging device which includes a first magnet system 2 for generating a steady magnetic field and also vaπous gradient coils 3 for generating additional magnetic fields which are superposed on the steady magnetic field and cause a gradient in the steady magnetic field in three respective orthogonal directions of a coordinate system X, Y, Z. The Z direction of the co-ordinate system shown corresponds by convention to the direction of the steady magnetic field in the magnet system 2. A measuπng co-ordinate system x, y, z (not shown) can be chosen independently of the X, Y, Z co-ordinate system shown in Fig 1 Generally speaking, a gradient in the x direction is referred to as a read-out gradient while a gradient in the y direction is referred to as a phase encoding gradient and a gradient in the z direction as a selection gradient. The gradient coils 3 are fed by the power supply unit 4. The MR device also includes an RF transmitter coil 5. The RF transmitter coil 5 serves to generate RF magnetic fields and is connected to an RF transmitter and modulator 6 A receiver coil is used to receive the magnetic resonance signal which is generated by the RF field in the object 7 to be examined in vivo, or in a part of the object to be examined, for example a human or animal body. The receiver coil may be the same coil as the RF transmitter coil 5. The magnet system 2 also encloses an examination space which is large enough to accommodate a part of the body 7 to be examined. The RF transmitter coil 5 is arranged around or on a part of the body 7 within the examination space. The RF transmitter coil 5 is connected, via a transmitter/receiver circuit 9, to a signal amplifier and demodulation unit 10 A control unit 11 controls the RF transmitter and modulator 6 and the power supply unit 4 so as to generate special MR imaging pulse sequences which contain RF pulses and gradients The phase and amplitude deπved from the demodulation unit 10 are applied to a processing unit 12. The processing unit 12 processes the received signal values, for example by way of two-dimensional Fouπer transformation, so as to form an MR image. An image processing unit 13 visualizes the MR image via a monitor 14.

The invention will be descπbed in detail hereinafter on the basis of an EPI imaging pulse sequence as shown in Fig 2. Fig 2 shows a pulse sequence 100 which, according to a first version of the method, is used, for example to obtain an in vivo MR image of a part of the brain of the body to be examined, for example a functional MR image which utilizes the BOLD effect in order to distinguish vaπous functional regions in the brain. In Fig 2 the time is plotted from left to πght and the vaπous rows indicate the temporal relationship between the RF pulses to be generated, the gradients to be applied and the MR signals to be measured. The upper row, denoted by the reference RF, shows the RF pulses to be generated and the three rows therebelow, denoted by the references G_x, G_y, G_z, show the gradients in the x direction, the y direction and the z direction, respectively. The row situated therebelow, denoted by the reference MR, shows the MR signals to be measured. The imaging pulse sequence 100 includes an excitation RF pulse 101, a refocusing RF pulse 102, a selection gradient 110, 111, an initial phase encoding gradient 120, phase encoding gradients 121 to 124, an initial read-out gradient 130 and read-out gradients 131 to 134. The imaging pulse sequence 100 commences with the generation of the excitation RF pulse 101. The flip angle of the excitation RF pulse preferably amounts to 90 degrees. The selection gradient 110 is applied simultaneously with the generation of the excitation RF pulse 101. The excitation RF pulse 101 thus excites spins in a selected slice of the brain of the body to be examined. Inter alia because of inhomogeneities in the steady magnetic field, the excited spins are dephased. The refocusing RF pulse 102 is generated a period of time tl after the generation of the excitation RF pulse 101. This refocusing pulse refocuses the excited spins so that at the instant 2.tl the rephasing of the spins is maximum and MR signals 140 to 143 can be measured. Furthermore, the initial phase encoding gradient 120 and an initial read-out gradient 130 are applied between the generation of the excitation RF pulse 101 and the generation of the refocusing RF pulse 102. The imaging pulse sequence 100 is repeated for various values of the initial phase encoding gradient 120, so that the MR signals 140 to 143 are measured along a first trajectory in the k-space, for example a k_x,k_y plane. In order to measure the MR signal 60, the read-out gradient 131 to 134 is applied after the generation of the refocusing RF pulse 102, phase encoding gradients which are referred to as "blips" then being applied at the zero crossings of the read-out gradient 131 to 134. The read-out gradient 131 to 134 is preferably a read-out gradient which alternates between two opposing directions. The number of MR signals 140 to 143 measured normally amounts to four. By repeating the pulse sequence 100 and applying the initial phase encoding gradients 120 with a different strength, subsequently a set of MR signals 140 to 143 is measured along the first trajectory which contains a number of lines, so that the first measuring plane is uniformly sampled. The first measuring plane contains, for example a rectangle or a square in the k_x,k_y plane of the k- space. The number of lines normally amounts to, for example 64 or 128. The lines extend parallel to the k_x axis.

In order to reduce the acquisition time, the first measuring plane is chosen so that it is asymmetrically situated relative to the center, the first measuring plane being subdivided into two sub-planes, for example by the k_x axis, the second sub-plane having a surface area which is smaller than that of the first sub-plane. The measured values are disposed in a matrix in positions which correspond to the first trajectory in the k_x,k_y plane. In order to render the matrix symmetrical, subsequently a number of values are added to the matrix. These. values correspond to non-measured points in the k_x,k_y plane and are estimated from the complex conjugate of the previously measured values of the MR signals 140 to 143 associated with the k values of the first trajectory in the first sub-plane, said k values being mirrored relative to the k_x axis. Subsequently, an MR image of the selected slice of the body to be examined is reconstructed, for example by means of two-dimensional Fourier transformation. In order to enhance the 7₂ ^* contrast in the MR image, according to the invention the first trajectory completed in the k_x,k_y plane during the acquisition is chosen to be such that a distance between a starting point of the first trajectory and the center of the k_x,k_y plane is greater than a distance between an end point of the first trajectory and the center of the k_x,k_y plane. The first trajectory along which the first measuring plane in the k_x,k_y plane is scanned will be described in detail with reference to Fig. 3. Fig. 3 shows the first measuring plane in the k_x,k_y plane, said first measuring plane being asymmetrically situated relative to the k_x axis. A first trajectory commences immediately after excitation; this trajectory represents the time integral of the applied initial phase encoding gradient 120 and the initial read-out gradient 130 in the k_x,k_y plane in a point A which is situated approximately at the center of the k_x,k_y plane. The k_x and k_y values of the time integral are calculated by means of

where γ represents a gyromagnetic constant,

G_yn represents the phase encoding gradient,

G_x represents the read-out gradient, t represents the time, and t' represents the time interval between the instant at which the MR signal is measured and the instant t=0 at which the excitation RF pulse is generated. After application of the initial phase encoding gradient 120 and the initial readout gradient 130, the time integral of the applied gradients increases to the value B. Subsequently, the refocusing RF pulse 102 is generated so that the starting point C is reached. Subsequently, the first trajectory CD, ... KL is completed within the first measuring plane by application of the read-out gradient 131 to 134 and the phase encoding gradients 120 to 124; preferably, the MR signals are then measured only during the completion of the sub- trajectories CD, ..., KL. The acquisition commences with the measurement of the first MR signal 140 near the starting point C of the first trajectory and ends with the measurement of the last MR signal near the end point L of the first trajectory. As a result of this choice of the starting point C and the end point L of the first trajectory, after the first trajectory has passed the center of the k_x,k_y plane no further MR signals are measured at the points in the k_x,k_y plane which are situated at a distance from the center of the k_x,k_y plane which is greater than the distance between the end point L of the first trajectory and the center of the k_x,k_y plane. As a result, an echo time TE, elapsing between the instant of application of the excitation RF pulse 101 and the instant at which the first trajectory passes the center of the k_x,k_y plane, is longer than an echo time T_E obtained by the method which is known from the cited United States US 4,767,991. Because the differences in the MR signals which are due to T_^' decay for different materials or tissue increase as a function of the length of the interval tl, the T_^' contrast in the MR image will be enhanced. Instead of subdividing the first measuring plane in the k_x,k_y plane into two sub- planes by way of the k_x axis, the k_y axis can also be used for subdivision. However, because the sampling of MR signals of a sub-trajectory AB, for example comprising 256 samples, requires approximately a period of 4 ms in the case of a sampling frequency of 64 kHz, a reduction of the number of profiles in the direction of the k_x axis is more effective so as to reduce the acquisition time. Such a trajectory in the k_x,k_y plane can also be used in other fast imaging pulse sequences such as, for example a known Rapid Acquisition or Recalled Echo (RARE) method, a known Gradient and Spin Echo (GRASE) method, or imaging pulse sequences utilizing helical trajectories in the k-space in order to sample the first measuring plane. Methods of this kind are known per se, inter alia from the handbook "Magnetic Resonance Imaging" as published by M.T. Vlaardingerbroek, Springer- Verlag, 1995, p. 117. In order to reduce the overall acquisition time for the MR signals even further while using the same echo time T_E, use can also be made of a PRESTO imaging pulse sequence. Such a PRESTO imaging pulse sequence is known from the cited article "A functional MRI technique combining pπnciples of echo- shifting with a train of observations (PRESTO)" by G. Liu et al., published in Magnetic Resonance in Medicine, 30, pp. 764-768 (1993) A PRESTO imaging pulse sequence of this kind will be descπbed with reference to

Fig. 4 shows a first and a second imaging pulse sequence TRl, TR2 of a

PRESTO imaging pulse sequence. Both imaging pulse sequences include an excitation RF pulse 401, 402, selection gradients 410 to 412, 413 to 415, phase encoding gradients 420 to 425, 426 to 431 and read-out gradients 440 to 445, 446 to 451. The imaging pulse sequence TRl commences with the generation of the excitation RF pulse 401, 402 which has a flip angle α Simultaneously with the generation of the excitation RF pulse 401, 402 the selection gradient 410, 413 is applied so as to excite spins in a slice of the body to be imaged The selection gradient 410, 413 is succeeded by the application of the equally large but oppositely directed first additional selection gradient 411, 414 Net dephasmg of spins excited (viewed in time) duπng a previous imaging pulse sequence is thus avoided. In order to measure MR signals 460 to 463 along a second trajectory in the k_x,k_y plane, thus sampling a second measuπng plane, the read-out gradients 441 to 444, 447 to 450 are applied after application of the selection gradient 411, 414 and an initial read-out gradient 440, 446 The second trajectory contains, for example a number of parallel lines which extend parallel to the k_x axis and regularly cover the second measuπng plane. The time integral of the initial read-out gradient 440, 446 and the read-out gradient 441 to 444, 447 to 450 according to the invention are chosen in such a manner that the spins are rephased at the middle of each of the four read-out peπods which coincide with the application of the read-out gradients 441 to 444, 447 to 450 Furthermore, at the end of each imaging pulse sequence TRl, TR2 there is also applied a second additional selection gradient 445, 451, so that the spins excited in the imaging pulse sequence TRl are rephased and yield an MR echo signal 460 to 462 in the second imaging pulse sequence TR2. Furthermore, an initial phase encoding gradient 420, 426 and further phase gradients 421 to 424, 427 to 430 are applied. Such further phase encoding gradients are referred to as "blips" and are applied at the zero crossings of the read-out gradients 441 to 444, 447 to 450 In order to reduce the acquisition time while maintaining the same echo time T_E, the second measuπng plane is chosen so as to be asymmetπcally situated relative to the center, the second measuπng plane being subdivided, for example by the k_x axis, into two sub-planes, the surface area of the second sub-plane being smaller than that of the first sub-plane. By application of the initial read-out gradient and the read-out gradients, the second trajectory completed in the k_x,k_y plane duπng the acquisition is chosen to be such that a distance between a starting point Cl of the second trajectory and the center of the k_x,k_y plane is greater than a distance between an end point LI of the second trajectory and the center of the k_x,k_y plane. Subsequent to the last blip 424, 430 in an imaging pulse sequence, an additional phase encoding gradient 425, 431 is applied in order to ensure that, at the end of an imaging pulse sequence TRl, TR2 and before a next excitation RF pulse 402, the phase of the spins is always the same in each imaging pulse sequence. In each subsequent imaging pulse sequence the initial phase encoding gradient 426 and the additional phase encoding gradient 431 are adapted, the magnitude of the "blips" being maintained so as to complete the second trajectory in the second measuring plane in the k_x,k_y plane. Fig. 5 shows an example of a second trajectory to be completed in the k_x,k_y plane.

The acquisition along the second trajectory in the second measuring plane Cl, Dl, Kl, LI starts at the starting point Cl and terminates at the end point LI. The measured values of MR signals 460 to 463 are disposed in a matrix in positions which correspond to the second trajectory in the k_x,k_y plane. In order to make the matrix symmetrical, subsequently a number of values is added to the matrix. These values correspond to non-measured points in the k_x,k_y plane and are estimated on the basis of the complex conjugate of the previously measured values of MR signals 140 to 143 associated with the k values in the first sub-plane in the k_x,k_y plane, being situated so as to be a mirror image relative to the k_x axis. Subsequently, an MR image of the selected slice of the object to be examined is reconstructed, for example by way of two-dimensional Fourier transformation.

Choosing such a second trajectory according to the invention may also be used in an echo-shifted Fast Field Echo method. A method of this kind is known from the article "A Fast Gradient Recalled MRI technique with Increased Sensitivity to Dynamic Susceptibility Effects" by C.T.W. Moonen et al., published in Magnetic Resonance in Medicine, No. 26, pp. 184-189, 1992. Such a method can be used, for example, to measure phase shifts in MR signals which are due to different temperatures within the object, thus enabling the reproduction of a two-dimensional temperature distribution of the object. The choice of the trajectory in the k-space according to the invention enables a reduction of the measuring time without the echo time being reduced.

Claims

CLAIMS:

1. A method for imaging an object by means of magnetic resonance (MR), which method involves an imaging pulse sequence which includes the following steps: generating an excitation RF pulse for the excitation of spins, measuring position-dependent MR signals during an acquisition by application of at least one read-out gradient along a trajectory in a measuring plane in a k-space, the trajectory containing a center of the k-space and the measuring plane being asymmetrically situated relative to the center of the k-space, characterized in that a distance between a starting point of the trajectory and the center is greater than a distance between an end point of the trajectory and the center of the k-space.

2. A method as claimed in Claim 1, wherein the k-space contains two orthogonal axes and one of the axes corresponds to the phase encoding gradient.

3. A method as claimed in Claim 1, wherein the k-space contains two orthogonal axes and one of the axes of the k-space corresponds to the read-out gradient.

4. A method as claimed in Claim 1, wherein the read-out gradient alternates between two opposed directions.

5. A method as claimed in Claim 1, wherein at least one refocusing RF pulse is generated subsequent to the excitation RF pulse in the imaging pulse sequence.

6. A method as claimed in Claim 1, wherein the imaging pulse sequence is repeated, a selection gradient being applied upon generation of the excitation RF pulse and a first additional selection gradient is applied after the selection gradient in order to avoid a net dephasing of the spins excited during a previous imaging pulse sequence, after which a second additional selection gradient is applied, after the measurement of MR signals, in the imaging pulse sequence in order to rephase the spins excited during the imaging pulse sequence.

7. A device for the magnetic resonance imaging of an object to be examined which is arranged in a steady magnetic field, which device includes:

- means for sustaining a steady magnetic field, means for applying gradients, - means for generating RF pulses to be applied to the object to be examined which is arranged in the steady magnetic field, a control unit for controlling the application of gradients and the generating of RF pulses and means for receiving and sampling magnetic resonance signals which are generated by pulse sequences which contain the RF pulses and the gradients, said control unit also being arranged to carry out a method which includes the following steps generating an excitation RF pulse, and measuring position-dependent MR signals during an acquisition by application of at least one phase encoding gradient and at least one read-out gradient along a trajectory in a measuring plane in a k-space, the trajectory containing a center of the k-space and the measuring plane being asymmetrically situated relative to the center of the k-space, characterized in that the control unit is also arranged to apply the at least one phase encoding gradient and the at least one read-out gradient in such a manner that a distance between a starting point of the trajectory and the center of the k-space is greater than a distance between an end point of the trajectory and the center of the k-space.