 |
Description  |
|
|
BACKGROUND OF THE INVENTION
The present invention relates to oxygen-17 nuclear magnetic resonance spectroscopy and imaging apparatus and methods in the living human, including the determination of blood flow and oxygen metabolism rates.
Various isotopes have been considered for use in connection with nuclear magnetic resonance (NMR) spectroscopy and imaging systems, which are in widespread use. Most NMR systems in use today are based on the hydrogen-1 isotope because that
isotope can be easily detected in humans for a number of reasons, including the fact that hydrogen-1 has a relatively large nuclear magnetic moment and the fact that its concentration in humans is relatively large. Those two factors allow conventional
NMR systems to generate NMR signals having a relatively large signal-to-noise (S/N) ratio. The S/N ratio is a measure of how easily an isotope can be detected in NMR.
Other isotopes have been considered for use in NMR. Research has been conducted in connection with oxygen-17 NMR, mainly in connection with small laboratory animals. It has been generally acknowledged that oxygen-17 NMR spectroscopy and imaging
are difficult for a number of reasons. First, oxygen-17 has a relatively weak magnetic moment, which is approximately 7.4 times weaker than the magnetic moment of hydrogen-1, which is commonly imaged by conventional NMR systems. Hydrogen-1 imaging is
commonly referred to as "proton" imaging. The weaker magnetic moment of oxygen-17 results in a smaller S/N ratio than that which can be accomplished with hydrogen-1.
Another obstacle to oxygen-17 NMR is the very low concentration, or natural abundance, of oxygen-17, which fact further reduces the S/N ratio which could be achieved in oxygen-17 NMR. The natural abundance of the oxygen-17 isotope in air is only
0.037%. Due to inhalation of air containing oxygen-17, the natural abundance of oxygen-17 in the water of tissues in animals and humans is also 0.037%.
These obstacles to oxygen-17 NMR have been recognized by those working in the field. For example, in U.S. Pat. No. 4,984,574 issued in 1991, Goldberg, et al. state: "It might, accordingly, be thought that existing NMR methods could be applied
to measure the oxygen content of living human fetuses. Such a direct application, however, seems infeasible for reasons including the following: First, naturally occurring oxygen consists mainly of .sup.16 O, whose nucleus possesses no magnetic moment
(hence has gyromagnetic ratio zero) and so cannot be studied by NMR. The natural abundance of .sup.17 O, which does possess a magnetic moment, is only 0.37% (sic, 0.037%) and its intrinsic sensitivity is approximately 1.08.times.10.sup.-5 times that of
.sup.1 H. As a result, the NMR signal from oxygen within a natural sample or living creature is only some four billionths as strong as the signal from an equal concentration of hydrogen nuclei within it, effectively ruling out any chance of detection by
available methods."
Instead of utilizing oxygen-17 NMR, oxygen-17 has been used as a contrast agent in hydrogen-1 NMR. For example, in Oxygen-17 Compounds as Potential NMR T.sub.2 Contrast Agents: Enrichment Effects of H.sub.2.sup.17 O on Protein Solutions and
Living Tissues, published in 1987, Hopkins, et al. disclose that the isotopic enrichment of solutions, living tissues, and organisms with oxygen-17 in the form of H.sub.2.sup.17 O shortens their proton (hydrogen-1) NMR transverse relaxation times
(T.sub.2) and suggest that oxygen-17 would therefore be useful as a contrast agent. In particular, Hopkins, et al. state: "Since changes in proton T.sub.2 can alter image intensity, localized variation in H.sub.2.sup.17 O concentrations could be
directly visualized as well as monitored in samples with the usual proton equipment and T.sub.2 procedures."
Oxygen-17 NMR research has been performed by Professor Gheorghe D. Mateescu and others at the Case Western Reserve University in Ohio. In an abstract entitled Oxygen-17 Magnetic Resonance Imaging from the Sixth Annual Meeting of the Society of
Magnetic Resonance in Medicine in 1987, Mateescu, et al. acknowledged the difficulties in oxygen-17 imaging: "Water is the primary signal source in Magnetic Resonance Imaging. So far, proton detection has been exclusively used because of its high
sensitivity, while the O-17 nucleus has generally been considered impractical for MRI, owing to its `unfavorable` properties: very low natural abundance, low detection sensitivity, and considerable quadrupolar broadening." The authors went on to state
that some of the "drawbacks" of oxygen-17 imaging could be turned into "advantages" and further stated: "The O-17 projection reconstruction of a T-shaped phantom shown in FIG. 1 compares favorably with the proton image of a similar phantom described in a
review by Andrew..sup.3 Although 1000 times less sensitive, the O-17 measurement in natural abundance takes only .about.10 times longer than the proton measurement. This is clue to its much faster (quadrupolar) relaxation time which allows many more
scans per unit time."
In Combined .sup.17 O/.sup.1 H Magnetic Resonance Microscopy in Plants, Animals and Materials: Present Status and Potential published in 1989, Mateescu, et al., in a section entitled "Basic Principles of Magnetic Resonance Imaging," state: "Since
the measurement is always made some time after pulse excitation and, in order to build sufficient signal-to-noise it is necessary to accumulate the signals of repetitive scans, the image intensity depends on both the T.sub.2 and T.sub.1 properties of the
specimen. This makes it possible to obtain T.sub.1 or T.sub.2 weighted images by selecting an imaging sequence with appropriately ordered and timed rf and gradient pulses." In connection with the magnetic resonance properties of oxygen-17, Mateescu, et
al. stated that the fast quadrupolar relaxation time of oxygen-17 allows pulsing rates at least 20 times faster without signal loss.
In an abstract entitled Oxygen-17 MRI and MRS of the Brain, the Heart and Coronary Arteries from the Eighth Annual Meeting of the Society of Magnetic Resonance in Medicine in 1989, Mateescu, et al. illustrated in vitro oxygen-17 images of a human
heart and coronary arteries. The images were of an excised human heart into which 20% oxygen-17 water was injected after ligation of the three ends of the coronary segment.
It has been suggested that oxygen-17 NMR spectroscopy and imaging might be used in humans. In In Vivo Measurement of Cerebral Oxygen Consumption and Blood Flow Using .sup.17 O Magnetic Resonance Imaging, published in 1991 by Pekar, et al., the
authors, who include the inventor of the present invention, state that "In summary, .sup.17 O NMR techniques have been used to measure cerebral blood flow and oxygen consumption in a 0.8-ml voxel in the cat brain. The technique has the potential to
image cerebral blood flow and oxygen consumption in humans."
Despite the above suggestions that oxygen-17 imaging might be successfully employed in humans, to the inventor's knowledge there has not been any report of successful oxygen-17 imaging in living human beings. Although the above abstracts and
papers suggest imaging could be accomplished in humans, the actual experiments were generally carried out with laboratory animals or inanimate objects, e.g. an excised in vitro human heart. The use of laboratory animals allows relatively large magnetic
fields, e.g. 9.4 Tesla device in the 1989 abstract of Mateescu, et al. referenced above, which are unsuitable for use on living human beings. The maximum magnetic field approved for use with living humans by the U.S. Food and Drug Administration is 2
Tesla. The use of a larger magnetic field for laboratory animals and inanimate objects results in a larger S/N ratio which allows images to be generated more easily.
Also, many of the oxygen-17 images reported above were generated based on the injection of relatively large amounts of oxygen-17 enriched water, the result of which increased the oxygen-17 concentration greatly above the 0.037 % natural abundance
level. Although high-concentration oxygen-17 injections increase the S/N ratio, such invasive methods of increasing the oxygen-17 concentration (which often resulted in the death of the laboratory animals) are inappropriate for use in living humans.
A prior art NMR imager which has been in widespread use for more than a year is the Signa 1.5 Tesla imager commercially available from General Electric. The GE imager has a superconducting magnet for generating a static magnetic field of 1.5
Tesla. The imager has various modes of operation including, for example, a multi-planar, gradient recalled (MPGR) mode.
In the MPGR mode, a resonant magnetic field and gradient magnetic fields are generated. The resonant magnetic field, which induces resonance of hydrogen-1 constituents, is generated via a number of Hermitian pulses transmitted at a repetition
rate TR. Although the RF frequency of the Hermitian pulses is variable between upper and lower limits as selected by the operator, the GE imager is typically used to cause magnetic resonance of hydrogen-1 constituents. Because the lower frequency limit
is too high, the GE imager is incapable of generating a Larmor frequency that would cause magnetic resonance of oxygen-17 constituents.
In the GE imager, the gradient magnetic fields are generated by a number of G.sub.x, G.sub.y, G.sub.z pulses. The spacing of the G.sub.x pulse with respect to the Hermitian pulse is determined by an echo time TE. In the GE imager, both the echo
time TE and the repetition time TR are selectable by the user within limits. The lower and upper limits for TE are 9 and 600 milliseconds (ms), respectively, and the upper and lower limits for TR are 34 and 6,000 ms, respectively. The GE imager also
performs sample averaging of a plurality of hydrogen-1 NMR signals for a particular area of a human body. The number of samples which are averaged, which is referred m as "NEX, " is selectable by the operator of the imager.
The GE imager is incapable of generating oxygen-17 images for at least the following reasons. The GE imager is incapable of transmitting an RF signal at a Larmor frequency for oxygen-17 constituents; the lower limits of the TE and TR times are
too high to facilitate the generation of oxygen-17 images; and the hardware of the GE imager does not provide sufficient low-noise amplification of the NMR signals to facilitate the generation of oxygen-17 images.
The accurate determination of the oxygen metabolism rate in various portions of a live human being is important for numerous clinical applications. For example, the cerebral oxygen metabolism rate is important for clinical applications include
assessing dementia, treating brain tumors, detecting cerebral ischemia (oxygen-deficiency), and understanding neurobehavioral disorders. The oxygen metabolism rates in other organs, such as the heart or lungs, is important for other clinical
applications.
One conventional manner of determining the rate of oxygen consumption in a live human being is positron emission tomography (PET). Such systems operate by detecting radioactive oxygen isotopes, such as oxygen-15. PET systems have a number of
significant disadvantages. Because PET systems rely on radioactive isotopes, they require cyclotrons, which are expensive and difficult to operate. Because the half-life of the radioactive isotope is typically short, e.g. 124 seconds of oxygen-15, the
cyclotron must be situated relatively near the PET system and provide an on-line supply of oxygen-15. As a result, PET systems are typically very expensive, on the order of $7 million. The cost of a single PET determination is also expensive, being on
the order of $10,000. A further disadvantage of the use of radioactive isotopes by PET systems is that PET determinations are usually not repeated more than two to three times in adults and are rarely used for children and infants. PET systems also
require catheterization.
Another significant disadvantage is the methodology on which PET systems are based. The consumption of oxygen results from metabolism of glucose in accordance with the following equation:
The fact that radioactive oxygen-15 is present in both the substrate (.sup.15 O labelled oxyhemoglobin) and the oxygen-15 water product, and the fact that the PET system detects both the radioactive oxyhemoglobin and radioactive oxygen-15 water,
make it difficult to distinguish between the oxyhemoglobin substrate and the water product as well as the reflow of oxygen-15 water and complicates and introduces errors in the calculation of the oxygen consumption rate. Although the PET system utilizes
semi-empirical equations to overcome that difficulty, the use of such equations casts doubt on the accuracy of the PET method to determine the rate of oxygen consumption. For example, it has been found that upon light stimulation, the metabolism of
glucose in an area of tissue increased while the rate of oxygen consumption in the tissue area decreased, which result is doubtful since glucose metabolism and oxygen consumption are directly, not inversely, related.
Mateescu, et al. have done NMR research in connection with inhalation of oxygen-17 by animals. In an abstract entitled Oxygen-17 Magnetic Resonance: In vivo Detection of Nascent Mitochondria Water in Animals Breathing .sup.17 O.sub.2 Enriched
Air from the 10th Annual Meeting of the Society of Magnetic Resonance in Medicine, Mateescu, et al. tracked the quantity of H.sub.2.sup.17 O in the head of a mouse. Mateescu, et al. stated that: "Our results show that good quantitation can be obtained
from volumes smaller than 1 cm.sup.3. This indicates that a good resolution should be obtained with larger animals and humans."
In the above abstract, although Mateescu, et al. disclose tracking the amount of H.sub.2.sup.17 O in the mouse head, they did not determine the true rate of oxygen consumption or the rate of blood flow in the mouse head because they failed to
account for the appearance of H.sub.2.sup.17 O due to reflow effects resulting from the recirculation of H.sub.2.sup.17 O-enriched blood. It has been recognized that the amount of H.sub.2.sup.17 O present in a portion of tissue is due to two components:
1) the production of H.sub.2.sup.17 O in the tissue due to metabolism; and 2) the change in H.sub.2.sup.17 O in the tissue due to H.sub.2.sup.17 O-enriched blood flow into and out of the tissue. Thus, Mateescu, et al. failed to account for the second
component above.
That the amount of H.sub.2.sup.17 O present in a portion of tissue is due to the above two components is described in In Vivo Measurement of Cerebral Oxygen Consumption and Blood Flow Using .sup.17 O Magnetic Resonance Imaging, published in 1991
by Pekar, et al., including the inventor. That paper describes a method of in vivo measurement of the rate of cerebral oxygen consumption and blood flow in a cat via oxygen-17 NMR via bolus injections of oxygen-17 enriched water and inhalation of
oxygen-17 enriched air. The increase of H.sub.2.sup.17 O in the cat brain due to reflow and the increase of H.sub.2.sup.17 O due to metabolism are shown in the graph of FIG. 3 of the paper.
There are several other methods used for the determination of localized blood flow, such as the magnitude of blood flow in the brain. One widely used method of determining the rate of localized blood flow in the human is the single photon
emission computed tomography (SPECT) system. However, SPECT systems have a number of significant disadvantages in that they use radioactive isotopes and they do not generate quantitative data regarding blood flow, but only generate data indicative of
the relative rate of blood flow in different portions of the body.
SUMMARY OF THE INVENTION
The present invention is directed to an apparatus for performing a method of generating oxygen-17 images of a body portion of a living person via magnetic resonance of oxygen-17 constituents. The method includes the steps of generating a static
magnetic field of a selected magnitude around the person, the selected magnitude not being substantially greater than 3 Tesla, and generating a resonant electromagnetic field of a selected frequency around the body portion. The frequency of the resonant
electromagnetic field and the magnitude of the static magnetic field are selected to cause magnetic resonance of oxygen-17 constituents in the body portion. The method also includes the steps of receiving oxygen-17 NMR signals generated from the
magnetic resonance of the oxygen-17 constituents and generating an image of the body portion based upon the oxygen-17 NMR signals.
The above method is capable of generating an oxygen-17 image even where the concentration of oxygen-17 constituents in the body portion is smaller than about 100 times the natural abundance of oxygen-17 in the body portion, and is even capable of
generating oxygen-17 images at the natural abundance level of oxygen-17 in the body potion.
In the method of imaging, electromagnetic field gradients having a magnitude not substantially greater than five gauss per centimeter may be generated around the person. Oxygen-17 images may be generated where the magnitude of the
electromagnetic field gradients does not exceed one gauss per centimeter. Higher field gradients result in higher resolution.
Averaging of a plurality of oxygen-17 NMR signals may be utilized to increase the S/N ratio. The resonant electromagnetic field may be generated by transmitting a plurality of RF pulses spaced apart by a repetition time TR less than 34
milliseconds to facilitate averaging of a relatively large number of NMR signals. The electromagnetic field gradients may be generated via a plurality of pulses having an echo delay time TE less than nine milliseconds to facilitate NMR signal averaging.
The apparatus of the invention also performs a method of determining the quantitative rate of blood flow in a body portion of a living person based upon oxygen-17 NMR signals generated while the person inhales a gas containing a concentration of
oxygen-17 different from the natural abundance of oxygen-17 in air. Preferably, the concentration of oxygen-17 in the inhaled gas is substantially greater than the natural abundance concentration, such as at least 10% oxygen-17. The method includes the
steps of generating a static magnetic field of a selected magnitude around the person and generating a resonant electromagnetic field of a selected frequency around the body portion. The frequency of the resonant electromagnetic field and the magnitude
of the static magnetic field are selected to cause magnetic resonance of H.sub.2.sup.17 O in the body portion. The method also includes the steps of receiving oxygen-17 NMR signals generated from the magnetic resonance of H.sub.2.sup.17 O in the body
portion during inhalation by the person of the oxygen-17 enriched gas and, based on the oxygen-17 NMR signals, generating a blood flow signal representative of the quantitative rate of blood flow through the body portion.
The method of determining blood flow, which is advantageously noninvasive, may determine the magnitude of organ tissue blood flow based on the difference between NMR signals generated during inhalation of the oxygen-17 enriched gas and an NMR
signal generated prior to inhalation of the oxygen-17 enriched gas. The method of determining blood flow may be repeated several times, in which case the blood flow magnitude is based on the difference between the NMR signals before and after each
period of enriched oxygen-17 inhalation.
The blood flow signal may be determined based upon the coefficients of an approximation curve, such as a least squares fit curve, of the NMR signals generated during inhalation. The approximation curve may be defined by a polynomial or an
exponential function. The blood flow signal may be determined based upon the concentration C.sub.b of H.sub.2.sup.17 O in the body portion and an arterial concentration C.sub.a of H.sub.2.sup.17 O, such as that of the aorta.
The apparatus of the invention also performs a method of determining the quantitative rate of metabolic oxygen consumption in a body portion of a living person based upon oxygen-17 NMR signals generated while the person inhales oxygen-17 enriched
gas. The method includes the steps of generating a static magnetic field of a selected magnitude around the living person and generating a resonant electromagnetic field of a selected frequency around the body portion. The frequency of the resonant
electromagnetic field and the magnitude of the static magnetic field are selected to cause magnetic resonance of H.sub.2.sup.17 O in the body portion. The method also includes the steps of receiving oxygen-17 NMR signals generated from the magnetic
resonance of H.sub.2.sup.17 O in the body portion during inhalation by the person of oxygen-17 enriched gas and, based on the oxygen-17 NMR signals, generating an oxygen consumption signal representative of the quantitative rate of metabolic oxygen
consumption in the body portion.
The method of determining metabolic oxygen consumption, which is advantageously noninvasive, may determine the magnitude of oxygen consumption based on the difference between NMR signals generated during inhalation of the oxygen-17 enriched gas
and an NMR signal generated prior to inhalation of the oxygen-17 enriched gas. The oxygen consumption signal may be determined based upon the coefficients of an approximation curve, such as a least squares fit curve, of the NMR signals generated during
inhalation. The approximation curve may be defined by a polynomial or an exponential function. The oxygen consumption signal may be determined based upon the concentration C.sub.b of H.sub.2.sup.17 O in the body portion and an arterial concentration
C.sub.a of H.sub.2.sup.17 O, such as that of the aorta.
The parameters to be determined by the present invention can be determined by either localized magnetic resonance spectroscopy (localized MRS) or magnetic resonance imaging. Variation in time of oxygen-17 localized MRS or MRI pixel intensity
during inhalation of natural air measures variations in regional oxygen metabolism rate and oxygen tissue blood flow due to physiological, neurological and psychiatric metabolic and circulatory effects on the regional rate of blood flow and oxygen
consumption.
These and other features and advantages of the present invention will be apparent to those of ordinary skill in the art in view of the detailed description of the preferred embodiment, which is made with reference to the drawings, a brief
description of which is provided below.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 illustrates an embodiment of an NMR spectroscopy and imaging system for use in connection with humans;
FIG. 2 illustrates a number of electrical signals generated during the operation of the NMR system of FIG. 1;
FIG. 3 is a block diagram of a first embodiment of the transceiver shown schematically in FIG. 1;
FIG. 4 is a circuit diagram of the preamplifier circuit shown schematically in FIG. 3;
FIG. 5 is a diagram of the T/R switch and RF coil shown schematically in FIG. 3;
FIG. 6 is an alternative embodiment of the transceiver shown schematically in FIG. 1;
FIG. 7 is a circuit diagram of the pulse converter circuit shown schematically in FIG. 6;
FIG. 8 is a circuit diagram of the transmit/receive switch shown schematically in FIG. 6;
FIG. 9 is a diagram of an alternative RF coil circuit;
FIG. 10 is an alternative embodiment of a preamplifier circuit;
FIG. 11a is a schematic diagram illustrating a number of organs in the human body;
FIG. 11b is a schematic diagram illustrating the components of H.sub.2.sup.17 O production in tissue in the human body;
FIG. 12 illustrates the concentration of H.sub.2.sup.17 O over time in a portion of tissue during inhalation of enriched oxygen-17; and
FIGS. 13-16 illustrate various methods of determining the oxygen-17 consumption and/or blood flow rates in a body portion of a human being.
DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT
FIG. 1 illustrates a nuclear magnetic resonance (NMR) spectroscopy and imaging system 10 which operates by magnetically resonating oxygen-17 constituents in the body portion of a living human being. The NMR system 10 includes a plurality of
elements for generating various magnetic fields to produce the magnetic resonance of the oxygen-17 constituents. These elements include a whole body, superconducting magnet 12 for generating a uniform, static magnetic field. The magnet 12 is referred
to as a "whole body" magnet since its circumference is large enough to accommodate the entire body of a person. The uniform static magnetic field is generated in the z-direction as shown in FIG. 1. Like most standard NMR systems currently in operation
in the United States, the superconducting magnet 12 generates a static magnetic field having a magnitude of 1.5 Tesla. While the magnet 12 is illustrated as a box 12 below the patient, it should be understood that the magnet 12 is generally cylindrical
in shape and surrounds the patient. As is conventional, a number of shim coils 13 provide additional static fields to supplement the main static field generated by the magnet 12.
The NMR system 10 includes a number of whole body, gradient field coils 14 which generate magnetic field gradients in the x, y, z directions in a conventional manner to facilitate NMR spectroscopy and imaging. In conventional 1.5 Tesla NMR
systems, the magnetic field gradients are typically between about 0.5-1.0 gauss/centimeter. The magnitude of the gradients is limited by FDA regulations. The gradient field coils 14 also surround the patient.
The NMR system 10 also includes an RF coil 16 for generating a resonant electromagnetic field at one or two radio frequencies for causing magnetic resonance of oxygen-17 constituents in the human patient. The resonant electromagnetic field may
be generated in any direction perpendicular to the static field. For example, if the static magnetic field is generated in the z-direction, the resonant field may be generated in any direction in the x-y plane. The RF coil 16 also receives electrical
NMR signals caused by the magnetic resonance of oxygen-17. While the RF coil 16 is shown schematically as a box 16 below the patient, it should be understood that the RF coil may surround the patient, or alternatively, local RF coils for generating
localized magnetic fields could be utilized, such as an RF coil for the head of the patient.
A power supply 17 supplies power to the magnet 12; the static fields generated by the shim coils 13 are controlled via a controllable power supply 20; a number of gradient field amplifiers 24 controls the fields generated by the gradient field
coils 14 via three lines 26, one line for each of the x, y, z directions; and a transceiver 30 controls the field generated by the RF coil 16 via a bidirectional line 32.
The elements 20, 24, 30 are connected to a conventional I/O interface 36 of a controller 40 via lines 42, 44, 46, respectively. The controller 40 also includes a processor 50, a random-access memory (RAM) 52, and a read-only memory (ROM) 54, all
of which are interconnected via a data bus 56. An image display device 58, such as a cathode ray tube (CRT), is also connected to the controller 40. Additional components, such as a boot and editing terminal 60, a tape drive 62 for storing data
relating to images, and a satellite station 64 for performing spectroscopy may also be connected to the controller 40.
During operation, the NMR system 10 generates a uniform static magnetic field via the superconducting magnet 12 and shim coils 13 and selectively generates the gradient and resonating magnetic fields via the coils 14, 16 so as to generate
oxygen-17 NMR signals corresponding to a selected vertical slice in the x-y plane of the human subject.
FIG. 2 illustrates one mode of NMR operation referred to as a multi-planar gradient recalled (MPGR) mode during which oxygen-17 images are generated. In general, the MPGR mode of operation for generating hydrogen-1 images is conventional and is
utilized in commercially available imagers such as the Signa 1.5 Tesla model commercially available from General Electric (hereinafter referred to as the "GE Signa").
Referring to FIG. 2, to cause magnetic resonance of oxygen-17 constituents, an RF signal is generated by the transmitter portion of the transceiver 30, G.sub.x, G.sub.y, G.sub.z signals are generated by the gradient field amplifiers 24 via the
three lines 26, and an NMR signal representing the magnitude of the oxygen-17 magnetic resonance is detected by the receiver portion of the transceiver 30.
The RF signal consists of a Hermitian pulse 70 with a magnitude envelope 72 having a central peak and a smaller peak on either side of the central peak. The central peak has a peak-to-peak magnitude of 140 millivolts (mv), and the width of the
central peak is 1.5 milliseconds (ms). The peak-to-peak magnitude of each smaller peak is 12 mv, and the width of each smaller peak is 0.85 ms. The RF frequency of the Hermitian pulse, represented by the oscillating lines inside the magnitude envelope
72, is based upon the Larmor equation set forth below:
where H is the magnitude of the static magnetic field, where .gamma. is the gyromagnetic ratio of the isotope to be magnetically resonated, and where f is the RF frequency necessary to cause nuclear magnetic resonance. To electromagnetically
resonate oxygen-17 in a static magnetic field of 1.5 Tesla, the RF frequency necessary is 8.66 MHz.
As is conventional, the G.sub.z signal has a positive slice-select pulse 76 which selects for magnetic resonance a particular x-y slice perpendicular to the z-axis, a negative re-phase pulse 78, and a positive killer pulse 80. The G.sub.y signal
has a positive pulse 82 which selects a particular y "line" in the x-y space to be magnetically resonated. The G.sub.x signal has a negative de-phase pulse 84 which, as is known, causes de-phasing of the precessing oxygen-17 constituents and a positive
readout pulse 86. An NMR signal 88 is generated during the readout pulses of the G.sub.x signal.
Generating the above sequence of pulses will generate NMR signals corresponding to one linear portion of an x-y plane. To get NMR signals for the entire x-y plane, the above sequence of signals is repeated a number of times using G.sub.y pulses
having different magnitudes. The signals other than the G.sub.y signal remain the same.
After an entire x, y plane is scanned, that same plane may be rescanned a number of times in order to generate plural sets of NMR signals. The plural sets of NMR signals may be averaged, the result of which affects the magnitude of the S/N ratio
of the NMR signals. When a relatively large number of samples are averaged, the S/N ratio increases due to the cancellation of noise in the NMR signal.
An important factor in facilitating the generation of oxygen-17 images is the timing at which the above pulses are generated and the rate at which the entire pulse sequence is repeated. The echo time, designated TE in FI | | |
