Driven Inversion Spin Echo Magnetic Resonance Imaging

1. Field of the Invention

This invention relates to improvements in inversion spin echo magnetic resonance imaging and, more specifically, it relates to a system which through control of certain perameters, will provide image contrast and feature detection which may be enhanced with respect to prior art practices while employing shorter scanning times.

2. Description of the Prior Art

The advantageous use of noninvasive and nondestructive test procedures has long been known in both medicine and industrial applications. In respect of medical uses, it has also been known that limiting a patient's exposure to potentially damaging x-ray radiation may advantageously be accomplished through the use of other noninvasive imaging procedures such as, for example, ultrasound imaging and magnetic resonance imaging. As to the latter, see, for example, The Fundamentals of Magnetic Resonance Imaging by Hinshaw et al. Technicare Corporation, 1984.

In a general sense, magnetic resonance imaging involves providing bursts of radio frequency energy on a specimen positioned in a main magnetic field in order to induce responsive emission of magnetic radiation from the hydrogen nuclei or other nuclei. The emitted signal may be detected in such a manner as to provide information as to the intensity of the response and the spatial origin of the nuclei emitting the responsive magnetic signal. In general, the imaging may be performed in a slice, or plane or multiple planes or three-dimensional volume with information corresponding to the responsively emitted magnetic radiation being received by a computer which stores the information in the form of numbers corresponding to the intensity of the signal. The Pixel value is established in the computer by employing Fourier Transformation which coverts the signal amplitude as a function of time to signal amplitude as a function of frequency. The signals are stored in the computer and may be delivered with or without enhancement to a video screen display such as a cathode-ray tube, for example, wherein the image created by the computer output will be presented through black and white presentations varying in intensity or color presentations varying in hue and intensity (and "saturation" or amount of "white" mixed in).

It has been known that magnetic resonance image intensity is dependent upon certain inherent physical properties of the tissues being investigated and timing intervals chosen by the user of the equipment. The physical properties of the tissues include the hydrogen density or density of the sensitive nucleus and two time factors which are known as T.sub.1 and T.sub.2. T.sub.1 which is also known as "T.sub.1 relaxation" is a measure of how long it takes the sample to regain its potential to produce a signal after a first pulse has caused it to respond to the pulsed RF excitation. This is sometimes considered as the time required to restore the longitudinal magnetization. T.sub.2 or "T.sub.2 relaxation" is a measure of the amount of time required for the magnetic resonance signal emitted by the radio frequency energy-excited proton to ideally dissipate to a point where it is generally imperceptible. At equilibrium, the transverse component of magnetization is at zero and the longitudinal component is equal to the initial magnetization. Decay to the former equilibrium is governed by the T.sub.2 relaxation and decay to the latter equilibrium is governed by the T.sub.1 relaxation. By properly selecting the timing intervals, the differences in hydrogen density or density of the sensitive nucleus (herein referred to as "N"), T.sub.1 and T.sub.2 values produce a difference in image intensity.

In general, in magnetic resonance imaging, it has been recognized that the differences in tissue T.sub.1 and T.sub.2 values are generally correlated such that an increase in one is accompanied by an increase in the other and a decrease in one is accompanied by a decrease in the other. Unfortunately, such parallel change does not contribute to a cooperative change in image intensity. For example, an increase in T.sub.2 causes an absolute increase in intensity while an increase in T.sub.1 usually decreases intensity. As a result, it is necessary to select the sequences so as to be dependent on either T.sub.1 or T.sub.2, but not both. The unfortunate consequence of such an approach is that many lesions or other elements desired to be visualized will go undetected if they only cause a significant change in one of the relaxation times T.sub.1 or T.sub.2. Also, competing T.sub.1 and T.sub.2 effects in respect of image intensity result in less contrast than would be the case if T.sub.1 and T.sub.2 effects were cooperative. Also, separate T.sub.1 sensitive and T.sub.2 sensitive pulse sequence acquisitions must be used to screen for both T.sub.1 and T.sub.2 changes, resulting in lengthy clinical studies.

It has been known to use spin echo and inversion spin echo techniques in magnetic resonance imaging. See generally pages 32 through 50 of the hereinbefore cited Hinshaw et al. publication.

In conventional spin echo imaging procedures, after an initial 90 degree pulse or general alpha-degree pulse, there are at predetermined intervals 180 degree RF pulses or magnetic field pulses which serve to refocus the transverse magnetization after the signal from the nuclei disappears to thereby cause the signal to reappear. These pulses which effect a spin echo will herein be referred to as "refocussing pulses". This regenerated signal is referred to a "spin echo". Depending upon whether the 180 degree pulses are in phase or 90 degrees out of phase with the 90 degree pulse, the resultant signals will either be solely positive or alternate between positive and negative. To the extent to which T.sub.2 relaxation has occurred prior to the generated spin echo, that portion of the signal is irretrievably lost.

In inversion recovery spin echo practices, there is an initial 180 degree pulse followed after a predetermined time period by a 90 degree pulse followed by a further time period and a repeat of the 180 degree spin echo cycle. This results in a potential for negative signals and a spreading out of the longitudinal magnetization thereby permitting images with greater dynamic range of contrast. The time between the first 180 degree pulse which inverts all magnetization and the subsequent 90 degree pulse which initiates the reading of the recovered longitudinal magnetization is deemed to be the "inversion time" which will be referred to herein as "TI".

It has been known in inversion-spin echo sequences of short TI intervals to provide additive T.sub.1 and T.sub.2 effects. See R. E. Steiner et al., Society of Magnetic Resonance in Medicine (1985), page 1208. The difficulty with this sequence is that it requires TI intervals which are much shorter than T.sub.1 so that the signals remain negative. As a result, T.sub.1 sensitivity is sub-optimal as it is maximal when TI equals T.sub.1 and contrast may be reduced even though T.sub.1 and T.sub.2 effects are additive. Moreover, the longitudinal magnetization recovers to nearly a zero magnitude during the short TI interval resulting in low signal to noise ratios.

It has been know to evaluate with mathematical models the desired pulse sequences and timing intervals in magnetic resonance imaging employing spin echo and inversion spin echo pulse sequences. See Mitchell et al. INVESTIGATIVE RADIOLOGY, Vol. 19, No. 5, pp. 350-360 (1984).

U.S. Pat. No. 4,254,778 discloses a system for driven equilibrium wherein after data collection, long TD intervals are sought to be avoided by applying a 90-180-90 triplet with the second 90 degree pulse shifted 180 degrees and the 180 pulses shifted 90 degrees in phase with respect to the first 90 degree pulse. This is said to drive the transverse magnetization at the spin echo peak back up to the positive longitudinal direction to thereby hasten recovery of longitudinal magnetization. This prior system has as its primary objective shortening of overall repetition time i.e., to produce maximal intensity in the shortest time. This is undesirable for certain types of imaging such as medical imaging as intensity is maximized at the expense of image contrast e.g., there is no sensitivity to differences in T.sub.1 as longitudinal magnetization is restored instantly regardless of T.sub.1 value. See also, Jensen et al., Medical Physics 14, 38-42 (1987) which discloses a 90-180-90 pulse triplet for saturating the signals outside of a plane. This triplet employs RF signals to drive the spin echo signal to the positive longitudinal direction rather than the negative direction of the present invention.

In spite of the foregoing, there remains a very real and substantial need to provide inversion spin echo procedures which will facilitate improved contrast with relatively short scanning periods.

As used herein, the terms "specimen" or "test specimen" shall refer to any object placed in the main magnetic field for imaging and shall expressly include but not be limited to members of the animal kingdom including humans; test specimens such as biological tissue, for example, removed from such members of the animal kingdom and inanimate objects which may be imaged by NMR or contain water or sources of other sensitive nuclei.

The present invention provides an improved method of inversion spin echo magnetic resonance imaging. A specimen is positioned within a main magnetic field and a source of RF signals in the form of pulses is positioned adjacent to the specimen. Receiver means for receiving signals emitted from the specimen responsive to the imposed RF pulses are positioned adjacent to the specimen and emit responsive output signals upon receipt of the specimen responses. Computer means receive the output signals from the receiver means and establish image information related thereto. The image information is provided directly or indirectly after storage to visual display means for displaying images from the image information received from the computer.

During an initial echo period, three RF pulses are imposed on the main magnetic field with the first and third pulses rotating the magnetization by a first value which may be 90 degrees, and the second pulse having a second value which is substantially larger than the first value and may be approximately double the first value e.g. 180 degrees. An echo is created with the second pulse and the third pulse converts this echo into negative longitudinal magnetization. After an inversion period and during a second echo period which follows, fourth and fifth RF pulses are imposed on the sample in the main magnetic field thereby creating a spin echo with the fifth pulse. The first through fifth pulses are preferably applied with an oscillating field generally perpendicularly to the main magnetic field. The receiver means receives the spin echo and the signal is converted to a digital signal by an analog-to-digital converter. The computer means receives the signal and responsive thereto establishes image information which is delivered to the visual display means in order to provide an image.

This cycle is preferably repeated a predetermined number of times in order to obtain the desired amount of data. In general, it is preferred that the image information be provided in accordance with a preferred relationship as set forth in Formula 2 hereinafter.

It is preferred that the initial echo period TE.sub.1 has a duration equal in time to the T.sub.2 relaxation period and that the second echo period TE.sub.2 has a duration substantially less than the the value of the T.sub.2 relaxation time. The inversion period preferably has a duration approximately equal to the value of the T.sub.1 relaxation time and the delay time TD for T.sub.1 recovery is preferably much greater than the value of the T.sub.1 relaxation period. It will be appreciated that use of a driven equilibrium after data collection at spin echo 36 can result in substantial shortening of the delay time TD.

The apparatus of the present invention employs magnetic field generating means for establishing a main magnetic field on a specimen and RF signal generating means for emitting pulsed RF signals to at least portions of the main magnetic field passing through the specimen. The pulse generating means has means for emitting three pulses during an initial echo period with the first and third pulses having a first value and the second pulse having a second value which is substantially larger than the first value and may be approximately double that of the first value in order to create with said second pulse an echo which is converted to negative longitudinal magnetization by the third pulse. After an inversion period during a second echo period, fourth and fifth pulses are imposed on the main magnetic field in the same sequence as the first and second pulses respectively in order to create a spin echo with the fifth pulse. In the event that the second and fifth pulses are magnetic gradient pulses the second value will have the effect of the second and fifth RF pulses. Receiver means receive signals emitted from the specimen responsive to the RF pulses and emit responsive output signals which in turn are received by computer means which establish image information related thereto. Visual display means display the images from the computer means image information.

In lieu of the second and fifth 180 degree RF pulses, refocussing gradient pulses may be employed.

It is an object of the present invention to provide an improved inversion spin echo magnetic resonance imaging system which facilities the use of reduced scanning times while providing good quality image contrast.

It is a further object of the present invention to provide a pulse sequence in which T.sub.1 and T.sub.2 sensitivity will be near maximal and additive for groups of tissues positively correlated with T.sub.1 and T.sub.2 values.

It is an object of this invention to provide such a system wherein parallel T.sub.1 and T.sub.2 changes produce parallel changes in pixel intensity.

It is a further object of this invention to provide such a system which has a desired signal to noise ratio.

It is a further object of the present invention to provide computer means which facilitate accomplishing the objectives of the invention.

It is a further object of the present invention to provide such a system which is compatible with existing equipment and is adapted to be economical to manufacture and use and may be used without significant retraining of individuals conducting the tests.

It is a further object of the present invention to provide such a system which is particularly adapted to provide improved contrast in magnetic resonance imaging of features in specimens.

These and other objects of the present invention will be more fully understood from the following description of the invention with reference to the illustrations appended hereto.

FIG. 1 is a schematic illustration of a magnetic resonance imaging system.

FIG. 2 is a representation of a conventional prior art inversion spin echo pulse sequence.

FIG. 3 is a representation of an RF sequence of the present invention and corresponding A/D converter and gradient values.

FIG. 4 is a schematic illustration of data generation and processing of the invention.

FIG. 5 is a flow diagram of a preferred form of software employable in the present invention.

A schematic illustration which presents the general concept of magnetic resonance imaging is shown in FIG. 1. An RF source 2 provides a pulse of radio frequency energy to the specimen in the form of a patient 4 placed in a magnetic field (which is on continuously). The specimen is generally aligned with the main magnetic field and the RF pulses are imposed perpendicular thereto. The main magnetic field is generated by magnetic field generator 6 and is generally perpendicular to the radio frequency (RF) field. This results in excitation of the nuclei within the area or volume to be imaged and causes responsive emission of magnetic energy which is picked up by receiver 8.

The receiver 8 may be a coil which has a voltage induced in it as a result of such responsive emission of magnetic energy. As a practical matter, separate coils or identical coils may be employed as the RF source 2 and the receiver 8. The signal emerging from receiver 8 passes through analog-to-digital (A/D) converter 10 and enters computer 12. Within the computer 12, the Fourier Transformations of signals converts the plot of amplitude versus time to a map of the distribution of frequencies by plotting amplitude versus frequency. In the computer, the Fourier Transformations are performed in order to establish the intensity values and locations of the specific pixels. These values may be stored, enhanced or otherwise processed and emerge to be displayed on a suitable screen, such as a cathode-ray tube 16.

In the conventional prior art inversion spin echo procedures, as is shown in FIG. 2, an initial pulse of 180 degrees is followed after elapse of an inversion period TI by a pulse of 90 degrees which, in turn, is followed by a pulse of 180 degrees which is midway between the 90 degree pulse and the echo as measured by time "TE". Time TD is the delay time for T.sub.1 relaxation after the spin echo and before the initiation of another cycle.

The present invention contemplates a major departure from the inversion spin echo magnetic resonance imaging concept. It contemplates the inversion pulse being a 90 degree pulse followed by a 180 degree pulse which, in turn, is followed by a 90 degree pulse. Such an inversion pulse triplet has been disclosed in connection with nuclear magnetic resonance spectroscopy and has been employed principally to correct for instrumentation imperfections, but not imaging. See M. H. Levitt et al., Journal of Magnetic Resonance, 33, 473 (1939); Edzes, J. Magnetic Resonance 17, 301 (1975); and Bydder et al., Journal of Computer Assisted Tomography 9, 659-675 (1985).

FIG. 3 illustrates a pulse sequence of the present invention which is designated as "driven inversion spin echo magnetic resonance imaging". Referring to FIGS. 1 and 3 and noting that the specimen in FIG. 1 is a patient, reference will be made to an arbitrarily oriented X-Y-Z gradient coordinate system and X'-Y'-Z' RF coordinate system which rotates about Z' at the frequency of the RF pulse (not shown). Z is normal to the selected slice or slices and Z' is along the main magnetic field. Assume imaging of a single transverse slice so that the patient has the longitudinal portions of his or her body aligned with the Z-axis along the main magnetic field. The Z-axis will also provide a Z-axis gradient which has planes of differing magnetic strength oriented generally perpendicularly with respect to the Z-axis and in relative spaced relationship. The Z-axis magnetic gradient serves to define the slices of the patient which will be imaged. The main magnetic field will cause nuclear magnetic moments of .sup.1 H nuclei (herein referred to as "protons") or other sensitive nuclei in the patient to be aligned parallel or antiparallel thereto in such a manner that the resultant spin vector is oriented generally parallel to the field's longitudinal axis. When pulses of RF energy of a frequency related to the resonant frequency associated with the main magnetic field are applied with an oscillating field perpendicular to the main magnetic field, resonance is established at that frequency so that energy is absorbed in the patient. The resultant spin vectors of the protons are caused to rotate from their orientation along the Z'-axis toward the X' or Y' axis. If the RF pulse is adequate to rotate the resultant spin vectors through 90 degrees, the pulse is designated a 90 degree pulse. If the RF pulse or general magnetic field pulse is adequate to rotate the resultant spin vectors through 180 degrees, the parent pulse is termed a 180 degree pulse. After removal of the RF pulse, T.sub.1 and T.sub.2 relaxation occur. A portion of the absorbed energy is emitted as a signal which can be detected by receiver means 8.

In addition to the foregoing phenomena, it is contemplated that pulsed magnetic gradients will be applied along the X and Y axes. It will be appreciated that references herein to any of the X-Y-Z or X'-Y'-Z' axes are solely for purposes of relative reference and that any of the coordinate axes may be designated as normal to the slice and any of the prime coordinate axes may be designated as the axis of the main magnetic field with resultant corresonding changes in identification of the other axes. Moreover, for oblique imaging the logical Z axis is defined to be normal to the slice. In this case the logical X, Y and Z gradients will each be madeup of mixtures of the physical X, Y and Z gradients. Hereinafter reference to X, Y and Z gradients refers to these logical gradients.

In a manner which will be described hereinafter, at certain times during the pulse sequence, it may be desirable to impose pulsed magnetic gradients on the X or Y axes.

An initial 90 degree pulse 20 is applied to a specimen placed in the main magnetic field. Subsequently, at a time equal to one-half echo time TE.sub.1, a refocussing pulse 24 which may be a 180 degree RF pulse is applied to the specimen. Pulse 24 creates echo 28 at a time period equal to one-half of TE.sub.1 thereafter. At the time of this echo, a further 90 degree pulse 26 is imposed and converts the echo 28 into inverted longitudinal magnetization. Time period TE.sub.1 is measured from the middle of pulse 20 to the middle of pulse 26 and is the time period between the initial 90 degree pulse and the creation of the echo 28 at pulse 26. Time period TI is the inversion time i.e., the time from the middle of the preparatory 90 degree pulse 26 to the middle of the 90 degree pulse 32 which begins the reading event or readout sequence. Subsequent to pulse 32 which is a 90 degree pulse, at a time equal to one-half echo time TE.sub.2, a refocussing pulse 34 which may be a 180 degree RF pulse is generated and thereafter, at a time equal to one-half TE.sub. 2, a spin echo 36 occurs and a signal is emitted responsive to the refocussing pulse 34. TE.sub.2 is the echo time from the middle of the initiating 90 degree pulse 32 to the peak of the resultant spin echo 36.

It will be appreciated that in the preferred practice the refocussing pulses 24 and 34 will be RF pulses substantially stronger and/or of longer duration than pulses 20, 26, 32 and 38, and preferably will be on the order of 180 degrees with the other pulses being 90 degrees. Pulse 20 may be in the range of about 70 to 110 degrees. Pulse 32 may be in the range of about 10 to 170 degrees if desired and refocussing RF pulses 24 and 34 can be replaced by refocussing gradient pulses.

Finally, after a period TD, a new cycle is begun by a 90 degree pulse 38. Time period TR is equal to the sum of TE.sub.1, TI, TE.sub.2 and TD. It will also be appreciated that multiple slices can be imaged by applying within this general sequence other RF and gradient pulses which are offset to affect and obtain information from other slices.

Several examples of the sequence in which the 90-180-90 pulse triplet of the present invention can be established will be considered. First of all, it is preferred that the triplet involves in-phase 90 degree pulses with a 180 degree pulse which is shifted 90 degrees out of phase and as a result, serving to drive the transverse magnetization down to the negative Z-axis direction. One sequence for pulses 20, 24, and 26 may have the 90 degree pulses imposed in a positive X'-direction, the 180 degree pulse imposed in a positive Y'-direction and the next 90 degree pulse imposed in a positive X'-direction. In the alternative, both of the 90 degree pulses can be in a positive X'-direction and the interposed 180 degree pulse in a minus Y' direction. Another approach would be to provide all of the pulses in an X'-direction with the first 90 degree pulse and the the 180 degree pulse being in a positive X'-direction and the second 90 degree pulse being in a minus X'-direction. A variation of this approach would be to have a 90 degree pulse in a positive X'-direction followed by a 180 degree pulse in the minus X'-direction and a 90 degree pulse in the minus X'-direction. It is assumed in these examples that the positive X'-direction is the direction of the first 90 degree pulse. The phases on pulses 32 and 34 can be arbitrarily set without adverse consequence. If in a given case the RF electronics and hardware allow arbitrary RF phases, the 90.sub.X' -180.sub.Y' -90.sub.X' triplet can be replaced by a 90.sub.X' followed by a 180 degree pulse with a phase alpha degrees away from Y', followed by a 90 degree pulse 2 alpha degrees away from X'. Similar variations can be made to the other three cases of RF phases set forth hereinbefore.

Considering further the sequence of pulses, 20, 24, 26, 32, 34 in FIG. 3, use of gradients in gaining information regarding spatial relationships will be considered. Gradients in the X and Y directions will generally be pulsed and be applied between pulse 32 and echo 36 and generally not during pulses 32 and 34. For example, a gradient along the X-axis may be imposed after termination of either 90 degree pulse 32 or 180 degree pulse 34. It will be appreciated that the RF pulse and gradient sequence provides a sequence which selects and images a slice or multiple slices of the specimen. As the pulse sequences of the present system are repeated, for example, 128 times, the strength of the gradients in the X and/or Y directions will be changed with successive cycles in order to provide meaningful information for imaging purposes. In general, within the gradient, the field strength will increase linearly with distance from the center. The nuclear moments within a slice of the specimen rotate in the transverse X'-Y' plane with a frequency which is proportional to the field strength within which they sit.

The X-axis may be employed as the frequency encoding axis from which signals relating to the high and low frequency portions may be provided in composite form and through Fourier Transformations be converted into a map of amplitude versus frequency. This permits a correlation between the X-coordinate of the signal and the frequency to be established.

Similarly, the Y-axis may be employed as the phase encoding axis. The varying moments under the influence of the Y-axis gradient can be employed to determine differences in gradient-induced phase oscillations between the high field region of the Y-gradient and the low field region of the Y-gradient. Fourier Transformations will provide the distribution of phase oscillation frequencies which can be related to positions along Y.

It will be appreciated that a principal feature of the present invention is the recognition of the numerous benefits which can be obtained from employing in inversion spin echo magnetic resonance imaging a pulse sequence of 90 degrees-180 degrees-90 degrees during an interval TE.sub.1 approximately equal to T.sub.2. Among the advantageous consequences of this approach are achieving cooperative T.sub.1 and T.sub.2 effects and obtaining T.sub.2 effects without decreasing signal to noise ratios. All of this results in providing better contrast between tissues, better signal to noise ratios and shorter imaging times which may be on the order of at least one-half the prior times required for imaging both T.sub.1 and T.sub.2 effects.

Referring again to FIG. 3, there is shown an analog-to-digital converter (A/D) which converts the echo signal 36 into a digital signal 40 of corresponding magnitude which is introduced into the computer 12 (FIG 1). As a result of the RF pulse sequence employed herein, parallel T.sub.1 and T.sub.2 changes will produce parallel changes in pixel and voxel intensity to thereby provide image information which will establish an image of desired contrast.

In the representation shown in FIG. 3, a slice of the patient has been imaged and the Z, Y and X gradients are represented by the lines AGZ, IGY and IGX.

The Z gradient 50 is on during the selective 90 degree pulse 20 to cause pulse 20 to excite only a slice and not the entire volume. Dispersion of phase normal to the slice caused by gradient pulse 50 is compensated for by gradient 52 and serves to keep the signal maximal. Gradient 54 serves to cause selective 180 degree pulse 24 to affect only a slice. Gradient 56 causes the second selective 90 degree pulse 26 to affect only a slice. Alternatively, the inversion pulse triplet can be nonslice selective in the case of the RF and gradient profile given by the dashed line in FIG. 3. Gradient 56 is preferably left on a little longer than gradient pulse 50 in order to dephase any remaining transverse magnetization. Gradient 58 causes 90 degree pulse 32 in the readout part of the sequence to affect only a slice. Gradient 60 is similar in function to gradient 52, but compensates for gradient 58. Gradient 62 is similar to gradient 54 but is for 180 degree pulse 34 in the readout part of the sequence.

On the Y gradient line, 64 is the phase encoding Y gradient which induces phase shifts in the spin echo signal thereby enabling spatial encoding in the Y direction. Gradient pulse 66 compensates for the phase dispersion caused by gradient 64. Gradient pulses 64 and 66 are stepped in strength for each repetition of the pulse sequence. For three dimensional spatial encoding there are also stepped Z gradient pulses analogous to 64 and 66 (not shown).

On the X gradient line, 70 is an X gradient pulse which compensates for the phase dispersion due to X gradient 72. Frequency encoding gradient 72 causes the spin echo signal to have a spectrum of frequencies enabling spatial encoding in the X-direction.

The purpose of the compensation gradients of FIG. 3 are to keep the signal as large as it can be. A gradient causes certain regions of a slice to have a transverse magnetization which processes or rotates at a high frequency, while other regions precess at a low frequency. As the NMR signal is derived from the entire slice or volume, the summed nuclear magnetization signal will consist of hydrogen nuclei which precess at a range of frequencies while the gradient is on. When the gradient is turned off, the hydrogen nuclei will again all be at the same frequency, but will not be aligned or in phase as those that moved faster during the gradient are now ahead of the slower ones. This causes destructive interference in the signal. The compensation gradient essentially "refocusses" this phase dispersion in much the same way that the 180 degree pulse does in a spin echo and may be substituted therefor. The gradient is applied with reverse polarity so that the part of the slice that precessed at high frequency now precesses at a low frequency and vice-versa. If both the positive and negative gradient pulses are of equal time duration such as 64 and 66 on the Y axis, then when the compensation gradient is turned off all of the hydrogen signals will be aligned or in phase. If there is a 180 degree pulse between the dephasing (principal) gradient and the rephasing (compensation) gradient then the compensation gradient must have the same polarity as the principal gradient as in 70 and 72 in FIG. 3. As a practical matter the duration and/or strengths of the compensation gradients are preferably tuned to maximize signal. In general, phase encoding gradient 64 can be on at any time between the 90 pulse 32 and the beginning of A/D conversion 40 excluding the time that the RF pulses are on. Similarly, the X compensation gradient 70 can be on any time from the 90 degree pulse 32 to the beginning of the readout gradient 72 excluding the time that the RF pulses are on. If 70 is on after the 180 degree pulse 34, it must be of reverse polarity.

The preferred formula (Formula 1) for the magnetic gradient establishment of pixel intensity is: ##EQU1## wherein: I=the actual pixel intensity

N=density of sensitive nuclear spins in the sample

K=a scale factor based on electronics, slice thickness and the like

This scale factor while not important for contrast serves to scale the entire image. It can be determined readily by measuring the signal (pixel) intensity of a "phantom" having a known spin density.

KN=the theoretical pixel intensity in the absence of RF pulses

T.sub.1 =T.sub.1 relaxation

T.sub.2 =T.sub.2 relaxation

TR=the repetition period for the entire cycle i.e. the sum of TE.sub.1, TI, TE.sub.2 and TD

TI=the inversion time i.e., the time from the middle of the second 90 degree pulse of the inversion pulse triplet that starts each sequence to the middle of the 90 degree pulse which begins the reading event

TE.sub.1 =the echo time from the middle of the initiating 90 degree pulse to the peak of the spin echo which is converted to inverted longitudinal magnetization

TE.sub.2 =the echo time from the middle of the 90 degree pulse beginning the reading event to the peak of the resultant spin echo.

In using this gradient pulse approach, the magnetic gradient pulses would be substituted for the RF pulse 34 and if desired the RF pulse 24 of FIG. 3. The magnetic gradient pulses if employed at 24 or 34 or both have the effect of a corresponding RF pulse. This Formula 1 also presents an approximation of the RF pulse approach, using 180 degree pulses at 24 and 34.

FIG. 4 shows in greater detail the handling of the signals in the current system. The magnetic resonance imaging equipment hardware 100, which may be of any desired form, cooperates with and is responsive to the pulse programming software 104 to generate the desired sequence of RF pulses and gradient pulses 108 as shown, for example, in FIG. 3. The signals, which have been digitized by the A/D converters, are then stored in a segment of the computer 110 and the scanning is repeated any desired number of times such as 128 times, for example, after which the raw data 112 is subjected to reconstruction through Fourier Transformations to create a map of frequency distribution with the two dimensional image 114 composed of individual pixels, for display.

A general relationship (Formula 2) of this invention which expresses a signal or pixel intensity of this system is: ##EQU2## wherein: I=the actual pixel intensity

N=density of sensitive nuclear spins in the sample

K=a scale factor based on electronics, slice thickness and the like

This scale factor while not important for contrast serves to scale the entire image. It can be determined readily by measuring the signal (pixel) intensity of a "phantom" having a known spin density.

KN=the theoretical pixel intensity in the absence of RF pulses

T.sub.1 =T.sub.1 relaxation

T.sub.2 =T.sub.2 relaxation

TD=the delay time from the peak of the acquired echo to the initiation of a new cycle initiated by a 90 degrees pulse i.e., the delay time necessary for T.sub.1 recovery in order for another cycle to be initiated

TI=the inversion time i.e., the time from the middle of the second 90 degree pulse of the inversion pulse triplet that starts each sequence to the middle of the 90 degree pulse which begins the reading event

TE.sub.1 =the echo time from the middle of the initiating 90 degree pulse to the peak of the spin echo which is converted to inverted longitudinal magnetization

TE.sub.2 =the echo time from the middle of the 90 degree pulse beginning the reading event to the peak of the resultant spin echo.

It will be appreciated from this relationship that the sign of the e.sup.-.spsp.TE.sup.1.spsp./T.sup.2 and e.sup.-.spsp.TI/T.sup.1 terms make it apparent that T.sub.2 sensitivity during the TE.sub.1 interval cooperates with the T.sub.1 sensitivity during the TI interval. It will be appreciated that T.sub.2 sensitivity will be maximal and cooperative with TE.sub.1 is approximately equal to T.sub.2 and TE.sub.2 is much less than T.sub.2 while T.sub.1 sensitivity will be maximal and cooperative when TI is approximately equal to T.sub.1 and TD is much greater than T.sub.1 and preferably about 2.5 to 4.0 times T.sub.1. TE.sub.1 may be about 0.25 to 4.0 times T.sub.2 and preferably about equal to T.sub.2. TE.sub.2 is preferably less than about 0.25 times T.sub.2.

Referring specifically to FIG. 5, there is shown a flow diagram of a form of the sequence employable in the software of the present invention. The dispatch function DSPTCH 124 establishes storage locations for the spin echo data and the phase of the receive coil i.e., the orientation in which the coil will detect the magnetization. This function is contained within lines 181 to 199 of the software listing which is provided hereinafter. Dispatch 124 send a control to PS1 128 (line 203) if the signal is the first signal for a given phase-encoding step ("pass mode") or sends a control signal to ADI 130 (line 207) if the signal is a later signal to be averaged with the previous signals for a given phase-encoding step ("add mode").

The term "JUMP.revreaction. always causes a jump as in a "GoTo" instruction. The optional use of the term "Jumpt" is employed only in studies gated to motion of the heart. The EKG would send a trigger that starts a sequence. The system may also be employed in a respiratory gating mode. SRPSN 132 receives signals from PS1 128. SRPSN 132 is employed for data acquisition in the pass mode. This codes for the saturation recovery or readout part of the sequence i.e., the entire TE.sub.2 interval. SRADN 134 is employed for data acquisition in the add mode. A control ultimately passes to SRCOM (line 362-371) which calls the SGLSLC MACRO 144 (line 311) which in turn calls the LINES MACRO 146 (line 312). The LINES 146 calls the routine ECHO1 (lines 325-330), if the first echo is to be generated or calls ECHOO 164 or ECHOE 168 (lines 333-338) if a later echo in a multiecho sequence is to be generated. ECHOO 164 is called for later odd numbered echos and ECHOE 168 (lines 341-347) is called for later even numbered echos. Referring to FIG. 3 once again, echo 36 would be called for at ECHO1 160 as this is the first and only collected echo.

The ECHO1 routine 160 sequentially calls SHPSR 180 (lines 255-287) which sets up the 90 degree pulse and the waiting period between the 90 degree pulse and the 180 degree pulse. Routine SHPSR 180 in turn calls PIPULS 182 (lines 709-730) which sets up the slice selection gradient 62 and the 180 degree pulse 34. The program then returns to POSTPI 183 (lines 484-511) which fills out the sequence after the 180 degree pulse up to before the readout gradient pulse 72 and which finally calls ACOXON 184 (ACQXON, line 511) which sets up the A/D converter and the readout gradient pulse and acquires the spin echo. SGLSLC 144 then calls FL 190 which calls for the TD interval, the driven inversion pulse triplet (lines 733-832), and the TI interval. The program then returns to DSPTCH 124 (line 180) for the next repetition (either the next phase-encoding step of a further signal-average of the same phase-encoding step).

In order to provide further guidance as to certain preferred practices of the invention, several examples of actual tests employing the invention will be considered. By way of comparing the timing of the present invention with the conventional types of inversion spin echo magnetic resonance imaging, two 128-step processes were performed wherein 128 cycles of data were gathered without signal averaging.

Employing the preferred relationship of the present invention as set forth in Formula 2, T.sub.1 was established at 700 milliseconds and T.sub.2 was established at 80 milliseconds. TE.sub.1 was established at 80 milliseconds, TI was 700 milliseconds, TE.sub.2 was 30 milliseconds and TD was 2800 milliseconds (four times T.sub.1). The total repetition time (TR) was 3610 milliseconds or 3.61 seconds. This equals approximately 7.7 minutes for a 128-step sequence which provides both T.sub.1 and T.sub.2 sensitivity.

By contrast, the conventional inversion spin echo magnetic resonance image would have TI, the time for the initial 180 degree pulse until the next 90 degree pulse at 700 milliseconds, TE the time from the 90 degree pulse to the echo at 30 milliseconds and TD the time from the echo to the next 180 degree pulse at 2800 milliseconds for a TR equal to 3530 milliseconds or 3.53 seconds. This converts to approximately 7.5 minutes for a 128-step sequence which provides only T.sub.1 sensitivity. A further sequence to obtain T.sub.2 sensitivity would be a spin echo sequence (90-180-echo-TD) repeat with TE=80 and TD=2800. The total repetition time (TR) was 2880 milliseconds or 2.88 seconds which equals 6.1 minutes for a 128-step sequence. The total time for T.sub.1 and T.sub.2 sensitive scans is 6.1 plus 7.5=13.6 minutes. This shows that the system of the present invention takes approximately one-half the time of the prior systems while providing equal or enhanced image contrast.

In order to disclose the best mode known to applicants for practicing the invention, a listing of the preferred software along with textual descriptions which indicate the significance of various segments of the software will be provided at this point. The computer means has logic means which controls the timing, signal processing, as well as operation of the magnetic fields and gradients and RF pulses. ##SPC1##

While it will be appreciated that modifications may be made to this software while still achieving the benefits of the present invention, this is the presently preferred approach.

As the specific hardware usable in the present invention may be of any conventional type previously known or subsequently generated for use in respect of inversion spin echo magnetic resonance imaging, disclosure of the same need not be provided herein in detail.

While readily known to those skilled in the art an example of hardware usable in the present invention will be provided. An Oxford Superconducting magnet in combination with a PDP-11/24 minicomputer and a sequencer computer that stores collected data before sending it to the PDP-11/24 minicomputer. Additional equipment may include a floating point system array processor, a PTS frequency synthesizer, DEC-VT-100 computer terminals, raster graphics display terminals and a matrix imaging camera for exposing paper and film.

While reference to visual display means has been made herein so as to suggest "real-time" display, it will be appreciated that visual display can be accomplished by storing the image information and subsequently displaying the same as on film and hard copies, for example. Such storage shall be deemed to be embraced within reference herein to visual display means, as an indirect means of getting information from the computer means to the visual display means.

While for convenience of disclosure herein reference has been made to collection of a slice or volume it will be appreciated that the system may be employed to obtain data from multiple slices simultaneously. This may be accomplished by pulsing different slices simultaneously by the use of appropriate gradients and RF pulse sequences during the TI and TD intervals.

It will be appreciated, therefore, that the present invention provides an effective means for increasing the speed of imaging while maintaining the same or providing enhanced image contrast and feature detection in inversion spin echo magnetic resonance imaging. All of this is accomplished in a reliable, economical and efficient manner.

Whereas particular embodiments of the present invention have been described above for purposes of illustration, it will be appreciated by those skilled in the art that numerous variations of the details may be made without departing from the invention as described in the appended claims.