CA2478466C - Squid detected nmr and mri at ultralow fields - Google Patents

Squid detected nmr and mri at ultralow fields Download PDF

Info

Publication number
CA2478466C
CA2478466C CA2478466A CA2478466A CA2478466C CA 2478466 C CA2478466 C CA 2478466C CA 2478466 A CA2478466 A CA 2478466A CA 2478466 A CA2478466 A CA 2478466A CA 2478466 C CA2478466 C CA 2478466C
Authority
CA
Canada
Prior art keywords
squid
nmr
sample
field
magnetic field
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CA2478466A
Other languages
French (fr)
Other versions
CA2478466A1 (en
Inventor
John Clarke
Robert Mcdermott
Alexander Pines
Andreas Heinz Trabesinger
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of California
Original Assignee
University of California
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of California filed Critical University of California
Publication of CA2478466A1 publication Critical patent/CA2478466A1/en
Application granted granted Critical
Publication of CA2478466C publication Critical patent/CA2478466C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00Arrangements or instruments for measuring magnetic variables
    • G01R33/20Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/28Details of apparatus provided for in groups G01R33/44 - G01R33/64
    • G01R33/32Excitation or detection systems, e.g. using radio frequency signals
    • G01R33/323Detection of MR without the use of RF or microwaves, e.g. force-detected MR, thermally detected MR, MR detection via electrical conductivity, optically detected MR
    • G01R33/326Detection of MR without the use of RF or microwaves, e.g. force-detected MR, thermally detected MR, MR detection via electrical conductivity, optically detected MR involving a SQUID
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00Arrangements or instruments for measuring magnetic variables
    • G01R33/02Measuring direction or magnitude of magnetic fields or magnetic flux
    • G01R33/035Measuring direction or magnitude of magnetic fields or magnetic flux using superconductive devices
    • G01R33/0354SQUIDS
    • G01R33/0356SQUIDS with flux feedback
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00Arrangements or instruments for measuring magnetic variables
    • G01R33/20Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/445MR involving a non-standard magnetic field B0, e.g. of low magnitude as in the earth's magnetic field or in nanoTesla spectroscopy, comprising a polarizing magnetic field for pre-polarisation, B0 with a temporal variation of its magnitude or direction such as field cycling of B0 or rotation of the direction of B0, or spatially inhomogeneous B0 like in fringe-field MR or in stray-field imaging

Abstract

Nuclear magnetic resonance signals are detected in microtesla fields.
Prepolarization in millitesla fields is followed by detection with an untuned dc superconducting quantum interference device "SQUID" (31) magnetometer.
Because the sensitivity of the SQUID is frequency independent, both signal-to-noise ratio "SNR" and spectral resolution are enhanced by detecting the NMR
signal in extremely low magnetic fields, where the NMR lines become very narrow even for grossly inhomogeneous measurement fields.

Description

= SQUID DETECTED NMR AND MRI AT ULTRALOW FIELDS
Government Rights The United States Government has rights in this invention pursuant to Contract No. DE-AC03-76SF00098 between the United States Department of Energy and the University of California.
Background of the Invention The invention relates generally to nuclear magnetic resonance (NMR) and magnetic resonance imaging (MRI), and more particularly to NMR and MRI at ultralow magnetic fields.
Nuclear magnetic resonance (NMR) is a technique for obtaining information about atoms and the molecules they form. NMR operates on atoms having nuclei in which at least one proton or neutron is unpaired. This imbalance causes these nuclei to spin on an axis like miniature tops and gives rise to a magnetic moment, i.e.
the nuclei behave like magnets with north and south poles.
When exposed to an external magnetic field, these spinning magnets attempt to align their axes along the lines of magnetic force. The alignment is not exact, however, resulting in a wobbly rotation (precession) about the force lines that is unique for each type of nuclei. If, while exposed to the magnetic field, the nuclei are bombarded with radio (RF) waves, they will absorb and re-emit energy at a specific frequency according to their rate of rotation. This resonating frequency therefore becomes a signature signal by which the nuclei can be identified.

When nuclei absorb the energy of an incoming radio wave, they are knocked out of alignment with the external magnetic field lines. As they subsequently lose this energy, the nuclei come back into alignment. The rate at which resonating nuclei realign themselves with magnetic field lines provides detailed information on their position and motion with respect to neighboring nuclei. This provides a noninvasive technique to study the structural, dynamic, and spatial relationships of atoms in a sample of molecules.
NMR has two basic subsets -- spectroscopy and imaging. In NMR spectroscopy, the frequency of the incoming radio wave is varied, and all of the different frequencies absorbed and emitted by the nuclei are measured to obtain a resonance spectrum. This NMR spectrum reveals the molecular makeup of the material down to the respective positions and motions of the constituent atoms.
In magnetic resonance imaging (MRI), the frequency of the incoming radio wave is kept constant, but the strength of the external magnetic field is varied.
The resulting signal corresponds to the total number of spinning nuclei present in any part of the sample, i.e. the atomic density of the sample at that point. Information obtained from an array of points can be translated by computer into a recognizable image.
Since the invention of MRI in the early 1970s, MRI scanners have steadily developed towards higher magnetic field strengths. The enhanced sensitivity attainable at high field makes it possible to resolve features at ever shorter length scales, and enables fast imaging experiments with close to real-time resolution. State-of-the-art clinical scanners operate at a field strength of 1.5 T, corresponding to a proton Larmor frequency of 64 MHz; currently, there is a drive to gain approval for 4 T imagers for clinical use. A
number of facilities around the world now have 7 T scanners for research purposes.
At the same time, the last three decades have seen continued effort toward the development of systems for MRI in low magnetic fields. Much of this work has been motivated by considerations of cost: a commercial full-body imager operating at 1.5 T
costs several million dollars, and the operation of such a machine places considerable demands on the infrastructure of the hospital or research facility. In addition, due to the size and complexity of the high-field system, it must necessarily remain fixed in one location, and the sample or subject must be transported to the system and inserted into the confining bore of the high field magnet; in certain cases this is simply not possible. A
low-cost, portable MRI scanner is extremely appealing, as is an open MRI
system, which would enable acquisition of MRIs at the same time that a medical procedure is performed. Inexpensive, portable imagers would enable MRI to address a wide variety of new problems, potentially transforming it from a highly specialized clinical and research technique to a much more widespread, flexible tool for rapid patient screening and general noninvasive imaging. However, any sort of portable or open MRI system would need to operate at relatively low magnetic field strengths.
Moreover, despite the serious disadvantage of reduced sensitivity, the images acquired in low field should, in principle, be of higher quality than those acquired in high magnetic field. An inevitable drawback of high-field imaging is that of susceptibility artifacts. When a heterogeneous sample is placed in a magnetic field, variations in magnetic susceptibility over the sample volume give rise to spurious magnetic field gradients. When these spurious gradients become comparable to the gradients that are used for encoding, the image is severely distorted. In medical imaging, the presence of dental fillings or jewelry is enough to destroy the MRI; abrupt changes in susceptibility at solid-liquid and solid-air interfaces inside the body, such as in the sinuses, produce distortions which are more subtle, but which nevertheless place strict limits on the achievable spatial resolution. Since the strength of the spurious gradients scales linearly with the strength of the applied field, it is possible to eliminate susceptibility-induced distortions entirely by imaging in low magnetic field.
Finally, T, contrast in tissue is enhanced in low magnetic field. Because of this, low-field images allow sharper differentiation of different organs and tissue types, and potentially contain richer information than the corresponding images acquired in high field. (It is interesting to note that, in the early days of MRI, many researchers were skeptical that high-field MRI would ever amount to a useful clinical tool, precisely because of the degradation of tissue contrast in high field.) There have been a number approaches to low-field MRI in recent years. These have generally relied on Faraday detection in a static field of order 10 mT to 100 mT, which is generated by an electromagnet. The main obstacle in these studies is the low sensitivity intrinsic to the low field experiment. In a different approach, H.C. Seton et al., "A 4.2 K receiver coil and SQUID amplifier used to improve the SNR of low-field magnetic resonance images of the human arm," Meas. Sci. Technol. 8, 198-207 (1997) employed a tuned SQUID magnetometer for NMR detection; the SQUID provided an SNR enhancement of a factor of 2.8 ¨ 4.5 over conventional detection in images acquired from room temperature samples in a field of 10 mT. In the low-field imaging work of A.
Macovski et al., "Novel approaches to low cost MRI," Magn. Reson. Med. 30, 221-(1993) and W. Shao et al., "Low readout field magnetic resonance imaging of hyperpolarized xenon and water in a single system," Appl. Phys. Lett. 80, 2032-(2002), spins were prepolarized in a field of 0.3 T, while the NMR signals were detected in a much lower field of 30 mT. Here, the homogeneity of the polarizing field was not crucial, and the prepolarization step led to an enhancement of sample magnetization by an order of magnitude. Using similar techniques, J. Stepignik et al., "NMR
imaging in the Earth's magnetic field," Magn. Reson. Med. 15, 386-391 (1990) and G. Planinsic et al., "Relaxation-time measurement and imaging in the Earth's magnetic field," J.
Magn.
Reson. Ser. A 110, 170-174 (1994), acquired MRIs in the magnetic field of the Earth (B,, ¨ 50 T), demonstrating the enhanced T, contrast attainable in low-field.
In both the works of Macovslci et al. and Stepignik et al., however, Faraday detection in the field of the Earth entailed substantial signal loss.
Superconducting Quantum Interference Devices (SQUIDs) are sensitive detectors of magnetic fields based on the quantum mechanical Josephson effect. SQUIDs are based on superconductors, whose resistance drops to zero when cooled to a critical temperature Tc. A SQUID is formed by separating its superconducting material with a very thin insulating barrier through which electron pairs can tunnel. This combination of superconducting material and insulating barrier forms a Josephson junction, i.e. two superconductors joined by a weak link. The SQUID consists of a superconducting ring or square interrupted in two spots by Josephson junctions. When sufficient electrical current is applied to the SQUID, a voltage is generated across its body. In the presence of a magnetic field, this voltage will change as the strength of the field changes.
Thus the SQUID turns a change in a magnetic field, which is more difficult to measure, into a change in voltage, which is very easy to measure.
For application purposes, SQUIDs are almost always coupled to auxiliary components. To form a magnetometer, a SQUID is connected to a flux transformer, a device consisting of a relatively large loop of superconducting material and a much smaller multiturn coil. Since the large loop picks up a magnetic field over a much greater area, the sensitivity of the SQUID to changes in magnetic field strength is boosted manyfold.
Originally SQUIDs were made with low Tc superconductors, e.g. niobium (Tc =
9.5K), which required cooling with liquid helium. More recently, high Tc SQUIDs have been made, using high Tc ceramic oxide superconducting materials, e.g. yttrium barium copper oxide (YBCO) materials (Tc = 93K), which only require cooling with liquid nitrogen, which is much less expensive and easier to work with than liquid helium. A
high Tc low noise SQUID is described in U.S. patent No. 6,023,161 issued 02/08/00.
Low transition temperature SQUIDs have been used experimentally to detect NMR and nuclear quadrupole resonance (NQR) signals, e.g. Dinh M. Ton That et al., "Direct current superconducting quantum interference device spectrometer for pulsed nuclear magnetic resonance and nuclear quadrupole resonance at frequencies up to 5 MHz," Rev. Sci. Instr. 67, 2890 (1996). Low Tc SQUIDs have been used to image polarized helium and xenon at relatively low fields, e.g. M.P. Augustine et al., "Low field magnetic resonance images of polarized noble gases obtained with a dc superconducting quantum interference device," Appl. Phys. Lett. 72 (15), 1908 (1998). The feasibility of using a high Tc SQUID to detect NMR signals has been demonstrated, S. Kumar et al., "Nuclear magnetic resonance using using a high temperature superconducting quantum interference device," Appl. Phys. Lett. 70 (8), 1037 (1997).
SQUIDs were first used in the 1980s to detect NMR signals in low magnetic field.
However, the majority of SQUID NMR studies have been performed on samples in the solid state, at liquid helium (LHe) temperatures. Recently, there has been increased interest in extending SQUID NMR techniques to samples in the liquid state, and in particular to systems which are biologically relevant. S. Kumar et al., "Broadband SQUID NMR with room temperature samples," J. Magn. Reson. B 107, 252 (1995) demonstrated NMR spectra from animal tissue measured at room temperature. H.C.

Seton et al., ibid., used SQUIDs to image room temperature samples in a field of 10 mT, and K. Sehlenga et al., "Low-field magnetic resonance imaging with a high-T, dc superconducting quantum interference device," Appl. Phys. Lett. 75, 3695-3697 (1999) and U.S. Patent 6,159,444 issued 12/12/00, used a SQUID magnetometer fabricated from the high transition temperature superconductor YBCO to image thermally polarized proton samples at room temperature in a field of 2 mT. Despite these early efforts, however, SQUID NMR studies of liquid samples remain extremely limited in number and in scope. The central challenge for SQUID NMR studies of liquids is that of low sensitivity. Thermal polarizations are two orders of magnitude lower at 300 K
than at 4.2 K. Moreover, in order to cool the SQUID below its superconducting transition temperature, it is necessary to thermally isolate the detector from the sample; filling factor is therefore quite low.
The NMR effect is produced by a spin magnetic moment on nuclei in a sample. A
magnetic field causes the spin magnetic moments to precess around the field at the Larmor frequency to which is proportional to the magnetic field.
In low field NMR (typically 5 10 ml) the spin precesses at correspondingly low frequencies, typically below 500 kHz, around the field direction. In conventional NMR, in which a resonant circuit is used to detect the precessing magnetization, the induced voltage signal V is proportional to the spin magnetization M and its rate of change (frequency) to, Since M is also proportional to the frequency to, V scales with of. As a result it is difficult to detect NMR signals at low fields with a conventional Faraday detector. In contrast, SQUIDs can be used to measure magnetic flux directly, resulting in much higher signal to noise (S/N) ratio at low frequencies. However, the use of SQUIDs for NMR/MRI has heretofore been limited, and not used at ultralow magnetic fields of tens or hundreds of microtesla.
Summary of the Invention Accordingly it is an object of some embodiments of the invention to provide method and apparatus for nuclear magnetic resonance and magnetic resonance imaging at ultralow magnetic fields.
In some embodiments, the invention is a method and apparatus for the detection of nuclear magnetic resonance (NMR) signals and production of magnetic resonance imaging (MRI) by obtaining NMR spectra of liquids in microtesla fields using prepolarization in millitesla fields and detection with an untuned dc superconducting quantum interference device (SQUID). Because the sensitivity of the SQUID is frequency independent, both signal-to-noise ratio (SNR) and spectral resolution are enhanced by detecting the NMR
signal in extremely low magnetic fields, where the NMR lines become very narrow even for grossly inhomogeneous measurement fields.
In some embodiments, the invention operates in ultralow magnetic fields, typically about 100-150 pT, but even tens of T and down to about 1 T or less, for detection and typically uses a field of about a few or tens of mT for prepolarization. Sample size can be very small but can be large, i.e. parts of the body. The detector is a SQUID magnetometer designed so =
that the SQUID detector can be very close to the sample, which is at room temperature.
In some embodiments, a cryogenic insert allows small separations between a sample maintained at room temperature and a SQUID detector operated in a liquid helium bath so that NMR signals from samples in the liquid state can be measured using a low-T, SQUID.
Prepolarization is performed by either resonant manipulation of spins in a magnetic field on the order of a few mT or non-resonant spin manipulation with switched static fields. The frequency-independent sensitivity of the untuned SQUID magnetometer allows measurement of NMR signals from the sample in extremely low magnetic fields, down to the order of I
T, where the proton Lannor frequency is of the order of tens of Hz. At these fields, NMR linewidths approach the lifetime limit, even in grossly inhomogeneous measurement fields. The reduction of signal bandwidth through reduction of the strength of the measurement field enhances both SNR and spectral resolution. While chemical shift information is lost in the low field, scalar couplings, which are field-independent, are preserved. These scalar couplings act as signatures of specific covalent bonds, Thus the invention includes a simple SQUID-based "bond detector", which yields accurate =
information about heteronuclear scalar couplings in microtesla fields.
Magnetic resonance imaging in ultralow magnetic field is based on the NMR at ultralow fields. In some embodiments, by exploiting the frequency-independent response of the untuned SQUID magnetometer, it is possible to enhance both the SNR and resolution (in this case, spatial resolution) of magnetic resonance images by reducing the strength of the measurement field to eliminate inhomogeneous broadening of the NMR lines.
According to one aspect of the present invention, there is provided a method for nuclear magnetic resonance (NMR) of a sample, comprising:
prepolarizing nuclear spins in the sample in a militesla magnetic field;
detecting nuclear magnetic resonance (NMR) signals from the sample in a microtesla magnetic field with an untuned low critical temperature (Tc) superconducting quantum interference device (SQUID) magnetometer.
According to another aspect of the present invention, there is provided apparatus for nuclear magnetic resonance (NMR) of a sample, comprising:
prepolarizing coils for providing a militesla magnetic field for prepolarizing nuclear spins in the sample; measurement coils for generating a microtesla magnetic field for detecting nuclear magnetic resonance (NMR) signals from the sample; an untuned low critical temperature (Tc) superconducting quantum interference device (SQUID) magnetometer for detecting nuclear magnetic resonance (NMR) signals from the sample.
Brief Description of the Drawings Figs. 1A-D illustrate the narrowing of signal bandwidth through reduction of measurement field strength.
Figs. 2A, B illustrate the process of MRI in microtesla magnetic fields.
Fig. 3 shows a cryogenic insert for SQUID-detected NMR spectroscopy of liquids.
Fig. 4 is a block diagram of the low-Tc SQUID spectrometer apparatus.
Fig. 5 shows an NMR pulse sequence for the apparatus of Fig. 4.

69666-101*
Figs. 6A, B illustrate SNR enhancement through reduction of the strength of the measurement field.
Fig. 7 is the NMR spectrum of 5 ml of 85% phosphoric acid (H3PO4) measured in a field of 2.6 pT.
Figs. 8A, B illustrate the resolution of scalar couplings in microtesla fields.
Fig. 9 shows the magnetic field and magnetic field gradient coils for SQUID detected MRI.
Figs. 10A, B are pulse sequences used for SQUID-detected MRI.
Fig. 11 illustrates a combined MEG and MRI apparatus.
Detailed Description of the Invention The invention is directed to certain improvements in NMR/MRI as described herein; other aspects of the NMR/MRI systems are conventional and not described since they are well known in the art.
A. Concept of Microtesla Field NMR
Line broadening due to field inhomogeneity is a major liability in liquid-state NMR. Spectral resolution, and therefore the information which one can extract about the interaction of nuclei with the local electromagnetic environment, is ultimately determined 8a by the width of the NMR lines: it is necessary that the strength of the interactions exceed the dispersion of Larmor frequencies in the sample. Moreover, for a fixed sample magnetization, the SNR achieved from a single FID or spin echo signal scales inversely with the inhomogeneously broadened width of the NMR line. For these reasons, high-resolution liquid state NMR requires exquisite field homogeneity. In a conventional high-field spectrometer, homogeneity is attained by supplementing the magnet with sophisticated and costly shim coils. In a state-of-the-art high-field system, homogeneity of a few parts per billion is achieved.
An alternate approach is simply to reduce the strength of the measurement field.
For a fixed relative field homogeneity, absolute homogeneity is enhanced by decreasing the strength of the measurement field. Of course, the sensitivity of a conventional NMR
detector falls off rapidly as field strength is reduced. However, this is not the case for an untuned SQUID magnetometer, which is sensitive to magnetic flux, rather than the rate of change of magnetic flux. In the context of NMR, this means that, for a fixed sample magnetization, the integrated intensity of the NMR signal ¨ the area under the NMR line ¨ is independent of the frequency of the NMR signal. This makes it possible to perform the following unconventional NMR process.
A sample of nuclear spins is polarized in a magnetic field of the order of 1 mT. In addition to the polarizing field, a much smaller measurement field is applied in an orthogonal direction. The polarizing field is then removed nonadiabatically, inducing precession in the much lower measurement field. The sample magnetization is fixed by the strength of the polarizing field. The width of the NMR line is determined by the absolute homogeneity of the measurement field. As the strength of the measurement field is reduced, the NMR line is compressed into a narrow band and the peak height grows, yielding an enhancement of both spectral resolution and SNR, as shown in Figs.
1A-D.
Fig. 1A shows a time trace of an NMR signal measured in a high inhomogeneous magnetic field. The dispersion of Larmor frequencies leads to rapid dephasing of the NMR signal. Fig. 1B shows the NMR spectrum corresponding to Fig. 1A. Fig. 1C
shows a time trace of an NMR signal measured in a weak magnetic field with the same relative homogeneity as that in Fig. 1A, using a detector whose sensitivity is independent of frequency. The NMR signal now appears at much lower frequencies. For a fixed sample magnetization, however, the amplitude of the NMR signal is unchanged.
Moreover, the effective spin-spin relaxation time T2* is much longer, as the absolute homogeneity of the measurement field has been enhanced by reduction of the measurement field strength. Fig. 1D shows the NMR spectrum corresponding to Fig. 1C.
The integrated intensity of the NMR signal is conserved upon reduction of measurement field strength. Therefore, as the NMR signal is compressed into a narrow band, the peak height grows, leading to an enhancement of both SNR and spectral resolution.
B. Concept of Microtesla Field MRI
There is an intimate connection between the inhomogeneously broadened width of the NMR line and the spatial resolution that can be achieved in an MRI
process. The important parameter is the NMR linewidth, which is determined by the absolute homogeneity of the measurement field, rather than the relative homogeneity.
Absolute field homogeneity is conveniently enhanced by reducing the strength of the measurement field. In the case of NMR detection with an untuned SQUID magnetometer, the reduction in measurement field entails no signal loss provided that the sample magnetization is fixed, for example by prepolarization in a higher field. The basic principles for SQUID-detected MRI in microtesla fields is then as shown in Figs. 2A, B. The sample of nuclear spins is polarized in field of order tens of mT, corresponding to a polarization of around 10-7. However, instead of detecting the NMR signal in a high field, where it is quite challenging to achieve narrow NMR lines, one measures the NMR signal in an extremely low magnetic field, where it is possible to approach the lifetime limit even for grossly inhomogeneous measurement fields. Now it is necessary to apply only modest magnetic field gradients to perform the encoding. As a result, the NMR signal is dispersed over only a narrow band. The NMR transients are therefore detected with a high SNR, and the time required to acquire the image is relatively short.
Fig. 2A shows an NMR spectra acquired from a sample (at top) consisting of two separate regions containing nuclear spins which is immersed in a magnetic field which is nominally homogeneous, but which involves spurious gradients which give rise to inhomogeneous broadening of the NMR lines. At high measurement field strengths, the absolute homogeneity of the field is relatively poor, and the NMR line is relatively broad (center). As the strength of the measurement field is reduced, the absolute homogeneity of the field is enhanced, and the NMR line is narrowed (bottom). In the case of fixed sample magnetization and detection with an untuned SQUID magnetometer, the reduction of measurement field strength also leads to an enhancement of SNR.
Fig. 2B is similar to Fig. 2A, but now a magnetic field gradient G, is applied in order to perform a one-dimensional MRI projection (top). At relatively high measurement field strengths, relatively strong gradients are needed in order to clearly resolve the two spatially distinct regions of the sample (center). As a result, the NMR signal is dispersed over a large band, and SNR is poor. At low measurement field strengths, inhomogeneous broadening of the NMR lines is largely eliminated. As a result, only modest magnetic field gradients are needed to resolve the two spatially distinct regions of the sample.
Therefore, the NMR signal remains confined to a relatively narrow band (bottom), and SNR is relatively high.
Even in the absence of homogeneous and inhomogeneous line broadening (infinite T2 and perfect field homogeneity), MRI resolution is ultimately limited by spin diffusion which occurs during the encoding intervals. However, the invention is directed at imaging at the mm length scale; therefore, spin diffusion does not present a problem.
Of course the effects of diffusion can also be overcome by limiting the length of the encoding intervals, and increasing the strength of the applied gradients.
C. Experimental Apparatus for NMR
1) Cryogenic Insert Fig. 3 shows a cryogenic insert 10 in a LHe filled dewar 11 for SQUID-detected NMR spectroscopy of liquids. A cell 12 containing the liquid sample was lowered into the tail section 14 of the insert and was maintained at room temperature by a resistive heater. The pickup coil 15 of the SQUID gradiometer, with inductance Lp, was wound around the tail section of the insert. Coils 16, 17 to produce the static magnetic field Bo (Bp) and excitation pulses B1 (Bm) were also located in the helium bath 18.
The dewar 11 was lined with a superconducting Pb shield 19 and the dewar was surrounded by a single-layer mu-metal shield 20 to attenuate the Earth's magnetic field and external magnetic disturbances.

In an illustrative embodiment, the inner compartment 21 of the insert, into which the sample was lowered, was 22 mm in diameter. This compartment was surrounded by a liquid nitrogen (LN2) jacket 22 with o.d. 100 mm to reduce the heat load of the insert on the LHe bath. At the 100-mm long insert tail section 14, however, the inner compartment extended directly into the LHe bath. A single continuous vacuum jacket 23 isolated the inner compartment of the insert from the LN2 jacket (and from the LHe bath in the tail region), and isolated the LN2 jacket from the LHe bath; the walls of the vacuum space were silvered, with a slit running the length of the insert. The separation between the sample space and the LHe bath was 5 mm. The insert 10 was surrounded by a number of styrofoam radiation baffles 24 which were covered with aluminum foil; these extended laterally from the body of the insert to the neck 25 of the LHe dewar, and served to reduce the heat load on the bath due to gaseous convection and direct radiation from the top of the dewar. To minimize the heat load due to thermal conduction of the Pyrex outer walls of the insert, the boiloff from the LHe bath was extracted from the brass top plate 26 of the insert, so that the insert was cooled by the evaporating helium gas. When the tail of the insert was not heated, the system consumed roughly 5 L of LHe per day.
2) SQUID Gradiometer and Readout A dc SQUID was fabricated using an all-liftoff Nb-A10x-Nb process. The SQUID parameters were: 2Ic - 5 A, R./2 - 10 Q, and L, - 350 pH. The peak-to-peak modulation of the SQUID was roughly 40 V when the SQUID was operated in a well-shielded environment. An 11-turn Nb input coil was integrated onto the SQUID
washer.
An illustrative SQUID pickup circuit was configured as a first-order axial gradiometer with pickup loop diameter 38 mm and a baseline of roughly 80 mm, and was wound from 3 mil Nb wire on a fiberglass form that fit around the tail of the cryogenic insert. For NMR measurements at low frequency, an untuned, or superconducting, input circuit was required. Superconducting contacts were made from the Nb wire pickup coil to the on-chip integrated Nb input coil.
A block diagram of the low-Tc SQUID spectrometer apparatus 30 is shown in Fig.

4. The dc SQUID 31 consists of a superconducting loop 32 interrupted by two Josephson junctions 33. When biased with a current Ib slightly above the critical current of the junctions, the SQUID acts as a flux-to-voltage transducer. To enhance its sensitivity to magnetic fields, the SQUID is often operated with a superconducting flux transformer 34 consisting of a pickup coil 35 tightly coupled to the sample and an input coil 36 tightly coupled to the SQUID loop 32. The flux transformer 34 operates on the principle of flux conservation in a superconducting loop, which involves no frequency dependence. Thus, the SQUID magnetometer can detect broadband at arbitrarily low frequencies with no loss in sensitivity.
The input coil 36 (with inductance Li) of the transformer 34 was integrated onto the SQUID chip; the niobium wire pickup coil 35 (with inductance Le) was wound in a gradiometric fashion around the tail section of a cryogenic insert. A single-layer solenoid 17 (Fig. 3) of copper wire wound on the sample cell produced the polarizing field (B, or Be). A set of coils 16 (Fig. 3) located in the helium bath provided the measurement field (Bo or B.). The SQUID was operated in a flux-locked loop with modulation at 2 MHz, and the signal from the SQUID was amplified, integrated, and fed back to the SQUID as a magnetic flux. The voltage across the feedback resistor Rf was thus proportional to the applied flux. In this way, the SQUID acted as a null detector of magnetic flux.
SQUID operating circuitry is known in the art. In the illustrative circuit of Fig. 4, the output of SQUID 31 is coupled through transformer 37 to amplifier 38 whose output is connected through feedback circuitry 42 to feedback coil 41. Circuitry 42 includes a lock-in detector 39 which receives inputs from amplifier 38 and oscillator 40.
The output of detector 39 is integrated by integrator 43. The outputs of integrator 43 and oscillator 40 are input into the two inputs of amplifier 44 which also has resistor Rf connected across the two inputs. it, is also connected to feedback coil 41. The output of amplifier 44 is input to computer 45 which controls reset circuit 46 which resets integrator 43.
The small-signal bandwidth of the 2 MHz flux-locked loop was around 700 kHz, and the slew rate greater than 10600/s. During spin manipulations, the feedback loop was disabled by shorting out the capacitor across the integrator. The signal from the flux-locked loop passed through a sample-and-hold stage (to remove the arbitrary dc level at the loop output) and a set of analog filters before digitization. Signal averaging was performed in software.

3) Static Field and Excitation Coils For NMR experiments involving resonant spin manipulation, the static Zeeman field was provided by a pair of coils located in the LHe bath and oriented orthogonally to the detection direction. These coils each consisted of 67 turns of Cu-clad NbTi wire wound on a 90 mm diameter frame; the separation of the coils was 55 mm. These coils provided roughly 1.2 mT per applied Ampere of current. Resonant pulses were provided by a pair of coils oriented along the detection direction and placed symmetrically with respect to the gradiometric pickup loops of the detector, in order to minimize the response of the SQUID to the excitation. Each coil consisted of 25 turns of insulated Nb wire wound on the 38-mm diameter frame on which the pickup coil was mounted. The excitation coils produced a field of roughly 830 ilT per applied Ampere.
For NMR experiments involving non-resonant spin manipulation, the 67-turn coils located in the helium bath were used to provide a measurement field of the order of a few microtesla. The spins were prepolarized along the detection direction in a field of a few millitesla; the polarizing field was provided by a one- or two-layer solenoid wound directly on the sample cell from Cu wire.
D. Microtesla Field NMR: Experimental In the experiments, the polarizing field was applied along the detection direction using a one- or two-layer solenoid of copper wire wound directly on the sample cell.
Currents of order 1 A were used to generate fields of order 1 mT. The measurement field was supplied by the 67-turn coils located in the LHe bath. The sudden switching of the polarizing coil induced magnetic transients that saturated the detector, giving a deadtime on the order of tens of milliseconds. Spin dephasing during this time resulted in signal loss that could be quite significant at higher measurement fields. A spin echo was used to refocus the sample magnetization. The echo was formed by reversing the direction of the measurement field, and thus the sense of precession of the nuclear spins.
A pulse sequence for microtesla field NMR is shown in Fig. 5. The polarizing field Bp of order 1 mT is applied for a time that is long compared to the spin-lattice time (T1) of the sample. Following nonadiabatic removal of the polarizing field, the spins precess in the measurement field Bm. As the spins precess, they lose phase coherence due to inhomogeneity of the measurement field. At a time t following the removal of the polarizing field, the direction of the measurement field is inverted.
Neglecting the effects of diffusion and in the absence of any background magnetic fields (fields generated by sources other than the current in the measurement coil), at each location in the sample, the spins see equal and opposite local magnetic fields before and following the inversion of the measurement field. Therefore the phase accumulated by each spin in the interval from t = 0 to t = t is cancelled by the phase accumulated in the interval from t = t to t =
Zr. At the time t = 2T, phase coherence of the spins is restored, and the echo amplitude is maximum.
The polarizing field Bp is applied along the detection direction, using a solenoid wound directly on the sample cell. The much weaker measurement field Bp, is applied in an orthogonal direction. The spins are polarized for a time that is long compared to the spin-lattice relaxation time T1. Precession is initiated by nonadiabatic turnoff of the polarizing field. A spin echo is formed by reversing the direction of the measurement field, and therefore the sense of precession of the nuclear spins.
Figs. 6A, B demonstrate the SNR enhancement achieved by reduction of the measurement field strength. Fig. 6A is an NMR (proton) spectrum of 5 ml of mineral oil acquired in a static magnetic field of 1.8 mT with homogeneity of roughly 10,000 ppm using a conventional Hahn spin echo sequence (7c/2-t-n-t-acq) involving resonant spin manipulation. Fig. 6B shows the NMR signal from the same volume of mineral oil, measured in a field of 1.8 1.1,T using the sequence of Fig. 5. The sample was polarized in a field of around 2 mT; the measurement field was applied with the same magnet used to acquire the spectrum in Fig. 6A. In this case, the proton resonance appears at 77 Hz. The sample magnetizations are the same in these two experiments; moreover, as the detector is untuned, the areas under the NMR lines are also the same. However, when the proton resonance was lowered from 77 kHz to 77 Hz, the NMR linewidth was compressed by a factor of 1000, and the peak height grew by the same factor. Reduction in the measurement field by a factor of 1000 therefore yielded an SNR enhancement of roughly 1000 ¨ note that Fig. 6A represents the average of 10,000 transients, while Fig. 6B was obtained from the average of 100 transients. In measurement fields of the order of 11.1T, SNR was a few tens without signal averaging from samples with a volume of a few milliliters and a polarization of order 10-8.
While there is no advantage in either resolution or SNR to be gained in reducing the measurement field past the point where the contribution of field inhomogeneity to the NMR linewidth becomes comparable to the natural linewidth, proton NMR signals wre measured at frequencies as low as 24 Hz.
E. Multinuclear Studies When the SQUID magnetometer is operated with an untuned input circuit, it detects broadband. Moreover, as the pulse sequence of Fig. 5 involves switched static fields rather than resonant spin manipulation, excitation occurs over a broad band. The technique is therefore ideally suited to studies of systems containing nuclei with different magnetogyric ratios, resonating at different frequencies. Fig. 7 shows simultaneous SQUID detection of 'H and 3113 resonances in a field of 2.6 jaT.
Fig. 7 is an NMR spectrum of 5 ml of 85% phosphoric acid (1131304) measured in a field of 2.6 T. The spectrum is the average of 1000 transients. The magnetogyric ratios of the spin-1/2 nuclei 'H and 3'P differ by a factor of 2.5. The proton resonance appears at 110 Hz; the 3'P resonance is clearly resolved at 44 Hz. The relative intensity of the two lines is determined by the different spin densities of the two nuclear species, as well as by the difference in thermal magnetizations brought about as a result of the difference in magnetogyric ratios.
F. Scalar couplings While all chemical shift information is lost in low magnetic field, scalar (or "J") couplings, which are field independent, are preserved. These scalar couplings act as signatures of specific covalent bonds. The enhanced resolution achieved by moving the resonance to low field makes it possible to accurately determine scalar coupling strengths, even in inhomogeneous measurement fields.
Fig. 8A shows the NMR spectrum obtained from a mixture of methanol and phosphoric acid, measured in a field of 4.8 T. The proton spectrum consists of a sharp singlet at 205 Hz. However, when the methanol and phosphoric acid are allowed to react to form the ester trimethyl phosphate, scalar coupling to the 3'P nucleus causes the proton resonance to split into a doublet, with a coupling strength J3[P,1-1] = 10.4 0.6 Hz which is characteristic of this particular next-next-nearest neighbor interaction (Fig. 8B). The proton doublet is easily resolved in a field of 4.8 T, despite a relative field homogeneity of roughly 10,000 ppm over the sample volume.
Fig. 8A is an NMR spectrum of 5 ml of 3 parts methanol, 1 part phosphoric acid (85% in water) measured in a field of 4.8 T. The spectrum is the average of transients. Rapid spin exchange with the protons in water obscures the proton-phosphorous scalar coupling in phosphoric acid, and the proton spectrum consists of a sharp singlet. Fig. 8B is an NMR spectrum of 3 ml of neat trimethyl phosphate (Sigma-Aldrich) measured in a field of 4.8 T. The spectrum is the average of 100 transients.
Electron-mediated scalar coupling of the nine equivalent protons to the 31I) nucleus splits the proton resonance into a doublet, with a splitting that is determined by the coupling strength J. For this particular coupling via three covalent bonds, J3[P,H] =
10.4 0.6 Hz.
Scalar coupling to the nine equivalent protons splits the 3113 resonance into ten lines; these are below the noise level.
Because electron-mediated scalar couplings between nuclear spins act as signatures of specific covalent bonds, these techniques could form the basis of a simple low-field NMR "bond detector", insensitive to chemical shifts, but yielding accurate information about scalar couplings. Such a detector could be applied to the study of analytes, chemical reactions, and molecular conformations. For example, the dispersion of 3-values for sp3 'H-'3C bonds is approximately 10 times greater than the NMR
linewidths achieved in our experiments. If the values of 3-couplings are known, then pure J-spectra could allow one to assign a number of molecular groups. Considering the highly developed techniques for isotopic labeling in biomolecular NMR, the use of such a method for following a "spy nucleus" through bond formation is an appealing possibility.
An illustrative proton spectrum was obtained from 5 ml of carbonyl labeled glycine in D20 in a field of 3.7 T. The resonance of the two equivalent a-protons is split into a doublet due to scalar coupling to the '3C nucleus. The coupling strength was determined from the lineshape of the doublet to be J2[C,F1] = 5 1 Hz.
G. Experimental Apparatus for MRI: Magnetic Field and Gradient Coils To perform MRI experiments in extremely low magnetic field, a convenient starting point was a zero magnetic field region. A set of three orthogonal cancellation coils 51 (Fig. 9) were used to zero out the field of the Earth over the measurement region.
These coils were wound around the perimeters of the six faces of a cube 50 that measured roughly 2 m on a side. Braces integrated into the cube structure supported a coil assembly 52 which consisted of: (1) a Helmholtz pair 53, used to produce a measurement field in the range of pa to tens ofilT; (2) a Maxwell pair 54, used to generate the diagonal component G. -d13./dz of the magnetic field gradient tensor; and two sets of saddle coils 55, 56 wound in the Golay geometry, used to generate the off-diagonal components G.
EdB./dx and G md13./dy of the gradient tensor. In addition, the braces supported the Y
fiberglass LHe dewar which housed the SQUID sensors. All support structures and coil forms were made from wood, which in many ways was an ideal material since wood is non-magnetic and non-conducting. The dimensions of the system were selected considering the eventual imaging of human subjects: an average-sized adult can fit (not too uncomfortably) into the measurement region at the center of the cube.
A schematic of the coil system 52 is shown in Fig. 9. Following the convention in the magnetic resonance community, the z axis lies along the measurement field direction;
the x axis is chosen to coincide with the vertical direction (the detection direction).
Fig. 9 shows the magnetic field and magnetic field gradient coils for SQUID-detected MRI. Six 100-turn coils 51 wound on each face of a cube, 2 m on a side, were used to cancel the magnetic field of the Earth. The measurement field Bo was produced by a Helmholtz pair 53 arranged in the center of the cube. The diagonal component G. of the first-order magnetic field gradient tensor was produced by a Maxwell pair 54; the off-diagonal gradients G. and Gy were produced by saddle coils 55, 56 wound in the Golay geometry (for each set of Golay coils, only two of the four saddle coils are shown). The detector was a second-order axial SQUID gradiometer oriented in the vertical (x) direction, and which was housed in a LHe dewar suspended in the center of the cube.
1. Cancellation Coils Each of the six cancellation coils consisted of 100 turns of 18 gauge copper wire wound in a groove cut along the outer edge of a wooden frame which made up one of the six faces of the cube. The coils on opposite faces of the cube were wired in series. Each pair had a resistance of roughly 30 52, and generated a magnetic field change of around 50 1AT per applied Ampere, so that only fractions of Amperes and modest voltages were required to cancel the three components of Earth's field. Heating in the cancellation coils was negligible. (For the record, a total of about 2.7 miles of 18 gauge Cu wire ¨
approximately 60 lbs ¨ went into the cancellation coils).
2. Measurement Field and Gradient Coils The measurement field was produced by a Helmholtz pair with radius 0.6 m arranged in the center of the cube. Each coil consisted of 20 turns of 18 gauge copper wire. The diagonal component G.of the gradient tensor was generated by a Maxwell pair with radius 0.6 m mounted outside the measurement field coils. Each coil consisted of 20 turns of 18 gauge wire. The off-diagonal gradients G. and Gy were generated by saddle coils wound in the Golay geometry from 20 turns of 22 gauge wire; the radius of curvature of the Golay coils was 0.6 m. These coils were wound in grooves cut along the outer edges of circular pieces of plywood. Per Ampere of applied current, these coils generated: 1) Bo = 30 T; 2) G. = 50 T/m; and 3) G. = Gy = 50 ptT/m.
3. Polarizing Coil In order to enhance the sample magnetization, the spins were prepolarized in a magnetic field of the order of tens of millitesla. Precession was induced either by non-adiabatic turnoff of the polarizing field (in which case the spins were polarized in a direction orthogonal to the measurement field), or by adiabatic turnoff of the polarizing field followed by resonant excitation in the much lower measurement field (in which case the spins were typically polarized along the measurement field direction). In the first instance, the polarizing field had to be switched in a time which was short compared to the Larmor period in the measurement field; in the second instance, the switching requirements were less strict: turnoff of the polarizing field needed to occur simply in a time which was short compared to the spin-lattice relaxation time T,. In either case, the requirement of rapid field switching, coupled with practical limits to the level of Joule heating which could be tolerated in the magnet wire, governed the design of the polarizing coil.

For rapid switching it is advantageous to use relatively heavy gauge wire and a relatively small number of turns. The polarizing magnet consisted of two poles wound from 18 gauge copper wire on fiberglass frames. On each pole, the magnet windings filled a 17 mm wide channel, with a 15 mm radius for the inner windings and a 70 mm radius for the outer windings. The separation of the two magnet poles was roughly 100 mm. A total of 510 turns was used for each magnet pole. The field at the center of the coils was approximately 4 mT per applied Ampere; the total resistance of the magnet pair was approximately 6 S2, while the inductance was roughly 100 mH. The heavy formvar insulation of the magnet wire seemed to tolerate Joule heating at a level of 100 W per magnet pole. To achieve higher polarizing fields, however, it would be necessary to water cool the magnet. While the homogeneity of the polarizing field was quite poor, this of course had no effect on NMR linewidth or SNR.
The condition for nonadiabatic turnoff of the polarizing field can be written as follows:
¨dB Bo) = yB2, dt where B is the instantaneous strength of the magnetic field. For measurement fields in the microtesla range, it is easy to meet the nonadiabatic switching condition, as the term on the right hand side becomes extremely small as the strength of the measurement field is reduced. Indeed, in our experiments, it was sufficient to drive the polarizing coil with a shaped voltage pulse from a Techron 7700 amplifier. At the turnoff of the pulse, the coil would discharge with an L/R time governed by the self-inductance of the coil (again, around 100 mH) and the output impedance of the amplifier (a few tens of ohms).
Driving the polarizing coil in this "controlled voltage" mode had the advantage that, following the turnoff of the polarizing field, a mechanical relay in series with the polarizing coil could be opened to decouple the SQUID gradiometer from noise generated by the Techron amplifier, as well as from the thermal noise of the polarizing coil itself.
With this simple switching scheme, the nonadiabatic criterion was satisfied for proton Larmor frequencies extending to around 2 kHz. As the strength of the measurement field was increased beyond this value, some loss in signal occurred, due presumably to adiabatic reorientation of the sample magnetization during turnoff of the polarizing field. At higher measurement frequencies, a scheme involving adiabatic removal of the polarizing field and resonant spin excitation was used.
4. Spin Echo Coil Resonant spin echo pulses were produced with a pair of coils oriented along the y direction. The coils each consisted of 10 turns of copper wire, wound on circular fiberglass forms with radius 60 mm which were mounted rigidly to the polarizing coil forms. In the case of the NMR and MRI experiments performed in this system, resonant echoes were preferable to echoes formed by inverting the direction of the measurement field, as the latter type of echo does not refocus dephasing from external magnetic field gradients (either the gradients Gz, Gy, Gz which were used for the encoding, or spurious gradients due to nearby magnetic objects). For Larmor frequencies of the order of 1 kHz, a typical TE pulse consisted of two or three cycles, and required currents of order 10 mA.
H. Experimental Apparatus for MRI: SQUID Sensors The SQUID system which was housed in the LHe dewar in the center of the cube consisted of four channels: a single large-area sensing channel and three orthogonal magnetometer references. The pickup coil of the sensor was configured as a second-order axial gradiometer oriented along the x axis (vertical direction); the gradiometer was therefore sensitive to the component c1213,/dx2 of the second-order magnetic field gradient tensor. The pickup loops of the gradiometer were wound from 3 mil Nb wire in grooves which were carefully machined in a cylindrical G10 fiberglass form. The radius of each pickup loop was roughly 15 mm, and the gradiometer baseline was 50 + 50 mm.
The form on which the pickup loop was wound fit snugly into the tail section of the fiberglass LHe dewar. The distance from the sensing loop of the gradiometer to the sample, which was positioned directly under the tail of the LHe dewar, was roughly 10 mm.
The sensing SQUID was housed in a superconducting box machined from a solid block of Pb and located in the LHe dewar at a distance of approximately 300 mm from the sensing coil of the gradiometer. Superconducting contact from the Nb wire pickup coil to the 11-turn Nb input coil integrated on the SQUID chip was achieved using the solder blob technique. The sensing SQUID was operated in a flux-locked loop with modulation at 2 MHz.

The NMR detector was a second-order axial gradiometer with an overall baseline of 100 mm. The probe also incorporated a three-axis reference magnetometer;
the reference SQUIDs were mounted on orthogonal faces of a G10 fiberglass cube which was positioned inside the cylindrical form on which the pickup coil of the gradiometer was wound.
The inductance of the sensing SQUID was roughly 350 pH. This yields for the 11-turn input coil a self inductance L, = 40 nH and a mutual inductance to the SQUID of M = 3.9 nH. For a 1 + 2 + 1 turn second-order gradiometer with pickup loop radius r =
15 mm, Lp = 700 nH. The above parameters yield a gradiometer sensing area A.
of approximately 3.7 mm2. Note that the sensing area of the gradiometer could be improved substantially by proper matching of Li and L. In particular, by increasing the number of turns in the input coil to around 45, so that L. - 700 nH, the sensing area of the gradiometer could be increased to roughly 8 mm2. However, even for a reduced sensing area of 3.7 mm2, system noise was dominated by external sources of thermal noise and interference, not by the intrinsic noise of the detector. In this case, inductance mismatch does not degrade SNR, as signal and noise are attenuated equally.
To measure the balance of the gradiometer with respect to uniform fields in the three orthogonal directions, known, highly uniform fields were applied to the sensor using the cancellation coils wound on the faces of the cube, while monitoring the flux that was coupled to the sensing SQUID. The typical gradiometer balance was one part in a few hundred for both in-plane and out-of-plane fields. To achieve this level of balance, however, parasitic inductance associated with the Nb wire leads which ran between the gradiometer loops, and which were not twisted together, was minimized.
Finally, the probe incorporated a three axis SQUID magnetometer consisting of three Nb-A10x-Nb SQUIDs mounted on three orthogonal faces of a G10 fiberglass cube.
This reference cube was mounted inside the fiberglass form on which the gradiometric pickup coil of the sensing SQUID was wound. Each of the reference SQUIDs had an effective area of roughly 0.03 mm2, and was operated in a flux-locked loop with modulation at 100 kHz. Analog subtraction of the reference signals from the gradiometer signal was used to improve the balance of the gradiometer, in an attempt to further reduce the contribution of distant noise sources to the system noise.

I. Choice of Measurement Field In all of the NMR and MRI experiments performed in this system, the sample magnetization was enhanced by prepolarization in a field of the order of tens of millitesla.
In the case of fixed sample magnetization and detection with an untuned SQUID
magnetometer or gradiometer, the integrated intensity of the NMR signal is independent of the strength of the measurement field. This allowed considerable freedom in the choice of measurement field strength for the experiments. The choice of measurement field was dictated by the following considerations: 1) inhomogeneous broadening due to spurious gradients generated by measurement field coil; 2) level of environmental interference and noise over the signal band; and 3) the strength of fields generated by gradient coils, relative to the measurement field strength (possible image distortion due to concomitant gradients).
J. NMR Experiments The procedure for performing NMR experiments in the cube was as follows.
First, appropriate currents were passed through the cancellation coils to zero out the static magnetic field of the Earth over the measurement region. For this purpose, a three-axis fluxgate magnetometer placed directly under the tail of the cryostat was used to monitor the field in the measurement region. The desired measurement field, of the order of microtesla or tens of microtesla, was then applied with the 1.2 m diameter Helmholtz pair. After allowing some time for the measurement field to stabilize, the sample was placed under the tail of the cryostat and tuned the SQUID gradiometer.
As mentioned above, two different polarization and excitation schemes were employed in the experiments: at lower measurement fields (corresponding to Larmor frequencies below about 2 kHz), spin precession was initiated by nonadiabatic turnoff of the polarizing field; at higher measurement fields, the polarizing field was reduced to zero adiabatically, and a resonant n/2 pulse was used to induce precession. In both cases, a spin echo, which was formed with a resonant TC pulse, was typically detected.
Figs. 10A, B show pulse sequences used for SQUID-detected MRI. In the sequence of Fig. 10A, precession is induced by the nonadiabatic turnoff of the polarizing field Bp (of the order of tens of millitesla), which is oriented in a direction orthogonal to the measurement field B.. The measurement field is static; a spin echo is formed with a resonant 71 pulse. In the sequence of Fig. 10B, the strength of the polarizing field Bp is reduced adiabatically to zero, and precession is induced with a resonant 71/2 pulse. In this case, the polarizing field is applied along the measurement field direction, to avoid signal loss due to imperfect fulfillment of the adiabatic switching criterion. In both cases, turnoff of the polarizing field is accomplished in a time of order 10 ms;
adiabaticity or nonadibaticity of the turnoff is determined by the Larmor period of the nuclear spins in the measurement field. The nonadiabatic turnoff was used for measurement fields below about 50 T; at higher measurement field strengths, the switching of the polarizing field was adiabatic, and only the adiabatic sequence was used.
In the experiments, polarizing field strengths were of the order of a few tens of millitesla, and the polarizing field was applied for a time which was long compared to the spin-lattice relaxation time T, of the sample (polarizing intervals of 2-3 s were used for water, and of around 100 ms for mineral oil). Following turnoff of the polarizing field and application of the echo pulse, mechanical relays in series with the polarizing and echo coils were opened. This was done to isolate the SQUID gradiometer from the noise of the amplifiers used to drive the polarizing and echo coils, as well as to prevent the flow of Nyquist noise currents in these coils. The flux-locked loop was enabled shortly after transmission of the echo pulse. The output of the flux-locked loop was passed through sample-and-hold and filtering stages and then digitized; signal averaging was performed in software.
When the external gradients were properly compensated, NMR linewidths of order 1 Hz were obtained from a test sample of water. With a polarizing field of roughly 40 mT, an SNR ratio of 50 was obtained in a single shot from a sample of around 20 ml of water.
K. MRI Experiments For imaging experiments, projection reconstruction was chosen over Fourier reconstruction due to the limited hardware requirements as well as the ease with which this encoding scheme is adapted to SQUID detection. MRIs were acquired from tap water or mineral oil phantoms. The phantoms were chosen to have translational symmetry in one direction, which was arranged to coincide with the axis of the second-order gradiometer (the x axis). The lateral (y-z) dimensions of the phantom were chosen to roughly match the dimensions of the sensing loop of the gradiometer (30 mm diameter).
The gradient component G. was nulled out using the appropriate set of Golay coils;
applied gradients Gy and G. were used to encode in the y-z plane. No attempt was made to perform slice selection, although of course the sensitivity of the SQUID
gradiometer fell off rapidly as a function of the separation of the source from the sensing loop.
Therefore the MRIs obtained were two-dimensional cross sectional images of the phantoms.
The first experiments were carried out with tap water phantoms. In this case, the long T, (of the order of seconds) made it possible to achieve very narrow lines of 1-2 Hz in low field, and therefore to take full advantage of signal bandwidth narrowing in order to enhance SNR and resolution. However, the long T, necessitated polarization times of the order of seconds. Due to the long polarizing interval, overall image acquisition times were rather long, of the order of a few hours.
Subsequent imaging experiments were performed with mineral oil, which in any case is a better match to human tissue in terms of its NMR relaxation properties. Here T2 is shorter, and (lifetime limited) proton linewidths of around 5 Hz were achieved. The broader NMR lines required encoding over a broader band in the MRI
experiments. On the other hand, the shorter T, of mineral oil enabled a significant decrease in the polarizing interval, to around 100 ms, resulting in a substantial reduction in image acquisition times.
Preliminary experiments such as this demonstrated the soundness of the concept of microtesla field MRI. However, the acquisition times could be substantially reduced by reduction of the system noise.
Further experiments have been done with a low-noise G10 fiberglass cryostat that can accommodate a gradiometer with pickup coil diameter up to 160 mm. A second-order gradiometer was constructed with an overall baseline of 150 mm, and pickup loop diameter 65 mm. The 1 + 2 + 1 turn pickup coil has an inductance of roughly 1.7 H, yielding a gradiometer sensing area A. = 7.6 mrn2. In addition, in order to eliminate the contribution to the system noise from distant sources of noise and interference, an eddy current shield surrounds the SQUID MRI system. The shield consists of a single layer of 1/8" (roughly 3 mm) 5052 Al plate (p = 4.8 S2 cm); its dimensions are 8' x 8' x 12'. At a frequency of around 5 kHz, the shield is roughly two skin depths thick, corresponding to an attenuation of interfering fields in the relevant frequency range by an order of magnitude. Inside this eddy current shield and with the new, large-area gradiometer, the thermal noise from the CTF dewar was around 2.5 fT/Hz''' at a frequency of 5 kHz; at the moment, the system noise is dominated entirely by thermal noise from the dewar. The current noise level of 2.5 fT/Hz'a is roughly a factor of 6 lower than the noise level achieved in the prior measurements, corresponding to a potential reduction in MRI
acquisition time by a factor of 36.
The larger pickup coil of the new gradiometer will allow acquisition of MRIs from larger samples, and some in vivo imaging experiments. However, the larger area of the sensing loop entails a significant loss of sensitivity to dipole sources (and in MRI, one is concerned with dipole sources). The signal flux coupled to a pickup loop of radius r from a dipole goes as r-' (this is seen easily if the problem is viewed in terms of mutual inductances). On the other hand, in the case of a system whose noise is dominated by the magnetic field environment, the noise flux coupled to the pickup loop will go as ?.
Therefore, in more than doubling the diameter of the gradiometer sensing loop, the sensitivity to magnetic dipoles decreased by more than a factor of 23 = 8, roughly canceling the advantage gained in reducing the magnetic field noise of the system. Due to these considerations, it will ultimately be desirable to replace the single large-area sensor with an array of smaller sensors, designed for maximum sensitivity to localized sources.
This is in fact the current trend in high-field clinical MRI, where system noise is dominated by inductively coupled losses from the human body.
L. Conclusion Thus the invention provides method and apparatus for obtaining nuclear magnetic resonance (NMR) spectra of liquids in fields of a few microtesla using prepolarization in fields of a few millitesla and detection with a dc superconducting quantum interference device (SQUID). Because the sensitivity of the SQUID is frequency independent, both signal-to-noise ratio and spectral resolution are enhanced by detecting the NMR signal in extremely low magnetic fields, where the NMR lines become very narrow even for grossly inhomogeneous measurement fields.
The technique of bandwidth-narrowing through measurement in microtesla fields can also be applied to magnetic resonance imaging (MRI). In MR', the ultimate spatial resolution is determined by the width of the NMR line in the absence of applied magnetic field gradients. For MRI performed in low field with linewidths approaching the lifetime limit, relatively high spatial resolution is achievable with modest magnetic field gradients. As a result the NMR signal is dispersed over only a narrow band, resulting in a high S/N ratio, and thus a short acquisition time. Furthermore, distortions due to spurious gradients generated as a result of spatial variations in magnetic susceptibility can be minimized in low field.
The invention thus leads to low-cost, portable MRI scanners as well as open MRI
systems, which would have significant advantages.
The low-field MRI techniques of the invention could readily be adapted for use with existing commercial SQUID systems designed to measure weak magnetic signals generated by electrical currents in the human body. Here, the frequency-independent response of the untuned SQUID magnetometer would be exploited to detect biomagnetic signals at frequencies in the range of Hz to tens of Hz, as well as to perform MRI at proton Larmor frequencies on the order of kHz. It would be particularly useful to develop a SQUID system which could be used for both MRI and magnetoencephalography (MEG). In MEG, an array of low-T, SQUID sensors (up to channels in state-of-the-art systems) is used to measure weak neuromagnetic signals (of the order of tens of if) in order to map out electrical activity in the brain;
the technique is rapidly becoming more widespread in clinical use, especially for studies of epilepsy and for pre-surgical screening of brain tumors. The ability to perform MRI would greatly increase the utility of the SQUID system, creating a versatile, new tool for studies of the human brain.
Fig. 11 illustrates a patient under a combined MEG/MRI helmet or apparatus 60.

Changes and modifications in the specifically described embodiments can be carried out without departing from the scope of the invention which is intended to be limited only by the scope of the appended claims.

Claims (27)

CLAIMS:
1. A method for nuclear magnetic resonance (NMR) of a sample, comprising:
prepolarizing nuclear spins in the sample in a militesla magnetic field;
detecting nuclear magnetic resonance (NMR) signals from the sample in a microtesla magnetic field with an untuned low critical temperature (Tc) superconducting quantum interference device (SQUID) magnetometer.
2. The method of Claim 1 wherein the detecting magnetic field is in the range of about 150 µT to about 1 µT.
3. The method of Claim 2 wherein the prepolarizing magnetic field is tens of mT.
4. The method of Claim 1 wherein the SQUID is maintained substantially at liquid helium temperature and the sample is at room temperature.
5. The method of Claim 1 wherein the prepolarization is performed by resonant spin manipulation and the prepolarization and detection fields are orthogonal.
6. The method of Claim 1 wherein the prepolarization is performed by nonresonant spin manipulation and the prepolarization and detection fields are parallel.
7. The method of Claim 1 wherein the detection magnetic field is substantially inhomogeneous.
8. The method of Claim 1 further comprising operating the SQUID in a flux locked mode.
9. The method of Claim 8 wherein the output of the SQUID is amplified, integrated, and then fed back to a flux modulating coil positioned near an input coil of the SQUID
10. The method of Claim 1 further comprising performing magnetic resonance imaging (MRI) of the sample by forming an image from the detected NMR
signals.
11. The method of Claim 1 further comprising applying encoding magnetic field gradients to the sample.
12. The method of Claim 1 further comprising simultaneously detecting multinuclear species.
13. The method of Claim 1 further comprising obtaining scalar ("J") coupling information from the NMR signals.
14. The method of Claim 1, wherein the SQUID comprises a superconducting input circuit.
15. The method of Claim 1 further comprising attenuating external magnetic fields including the Earth's magnetic field at the sample.
16. The method of Claim 15, wherein attenuating external magnetic fields comprises using cancellation coils to zero out such fields.
17. The method of Claim 15, wherein attenuating external magnetic fields comprises using a shielding material.
18. Apparatus for nuclear magnetic resonance (NMR) of a sample, comprising:
prepolarizing coils for providing a militesla magnetic field for prepolarizing nuclear spins in the sample;

measurement coils for generating a microtesla magnetic field for detecting nuclear magnetic resonance (NMR) signals from the sample;
an untuned low critical temperature (Tc) superconducting quantum interference device (SQUID) magnetometer for detecting nuclear magnetic resonance (NMR) signals from the sample.
19. The apparatus of Claim 18 further comprising a flux locked loop connected to the SQUID.
20. The apparatus of Claim 18 wherein the detecting magnetic field is in the range of in the range of about 150 µT to about 1 µT and the prepolarizing magnetic field is tens of mT.
21. The apparatus of Claim 18 further comprising a liquid helium dewar in which the SQUID is mounted to maintain the SQUID at liquid helium temperature while the sample is at room temperature.
22. The apparatus of Claim 18 further comprising means to perform magnetic resonance imaging (MRI) of the sample by forming an image from the detected NMR signals.
23. The apparatus of Claim 22 further comprising means to perform MRI
include gradient field coils.
24. The apparatus of Claim 22 further comprising a SQUID array for performing magnetoencephalography.
25. The apparatus of Claim 18 further comprising one or more cancellation coils to zero out substantially all of the Earth's magnetic field around the sample.
26. The apparatus of Claim 18 further comprising a magnetic shield material surrounding the dewar to attenuate external magnetic field including the Earth's magnetic field.
27. The apparatus of Claim 18, wherein the SQUID comprises a superconducting input circuit.
CA2478466A 2002-02-06 2003-02-06 Squid detected nmr and mri at ultralow fields Expired - Fee Related CA2478466C (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US35557702P 2002-02-06 2002-02-06
US60/355,577 2002-02-06
PCT/US2003/003712 WO2003067267A2 (en) 2002-02-06 2003-02-06 Squid detected nmr and mri at ultralow fields

Publications (2)

Publication Number Publication Date
CA2478466A1 CA2478466A1 (en) 2003-08-14
CA2478466C true CA2478466C (en) 2013-09-17

Family

ID=27734537

Family Applications (1)

Application Number Title Priority Date Filing Date
CA2478466A Expired - Fee Related CA2478466C (en) 2002-02-06 2003-02-06 Squid detected nmr and mri at ultralow fields

Country Status (12)

Country Link
US (5) US6885192B2 (en)
EP (1) EP1474707B1 (en)
CN (1) CN1287160C (en)
AP (1) AP2004003127A0 (en)
AU (1) AU2003210906B2 (en)
CA (1) CA2478466C (en)
EA (1) EA006832B1 (en)
IL (1) IL163669A0 (en)
MX (1) MXPA04008593A (en)
NZ (1) NZ534963A (en)
WO (1) WO2003067267A2 (en)
ZA (1) ZA200406984B (en)

Families Citing this family (96)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EA006832B1 (en) 2002-02-06 2006-04-28 Те Риджентс Оф Те Юниверсити Оф Калифорния Squid detected nmr and mri at ultraflow fields
US6724188B2 (en) * 2002-03-29 2004-04-20 Wavbank, Inc. Apparatus and method for measuring molecular electromagnetic signals with a squid device and stochastic resonance to measure low-threshold signals
US6995558B2 (en) * 2002-03-29 2006-02-07 Wavbank, Inc. System and method for characterizing a sample by low-frequency spectra
EP1511995A4 (en) 2002-04-19 2005-10-19 Wavbank Inc System and method for sample detection based on low-frequency spectral components
JP4221192B2 (en) * 2002-06-14 2009-02-12 株式会社日立ハイテクノロジーズ Biomagnetic measurement device
WO2006023208A2 (en) * 2004-07-22 2006-03-02 The Regents Of The University Of California Low field magnetic resonance imaging
EP1792179A2 (en) * 2004-07-27 2007-06-06 Nativis, Inc. System and method for producing chemical or biochemical signals
WO2006052236A1 (en) * 2004-11-03 2006-05-18 The Regents Of The University Of California Nmr and mri apparatus and method involving a squid magnetometer
US7187169B2 (en) * 2004-11-03 2007-03-06 The Regents Of The University Of California NMR and MRI apparatus and method
WO2008024092A2 (en) * 2005-06-17 2008-02-28 The Regents Of The University Of California Nmr, mri, and spectroscopic mri in inhomogeneous fields
US20070025918A1 (en) * 2005-07-28 2007-02-01 General Electric Company Magnetic resonance imaging (MRI) agents: water soluble carbon-13 enriched fullerene and carbon nanotubes for use with dynamic nuclear polarization
ZA200707480B (en) * 2005-10-06 2008-11-26 Tech Resources Pty Ltd Gravity gradiometer
US7573264B2 (en) * 2005-11-28 2009-08-11 The Regents Of The University Of California Atomic magnetic gradiometer for room temperature high sensitivity magnetic field detection
US7573268B2 (en) * 2006-02-22 2009-08-11 Los Alamos National Security, Llc Direct imaging of neural currents using ultra-low field magnetic resonance techniques
WO2008008447A2 (en) * 2006-07-12 2008-01-17 The Regents Of The University Of California Portable device for ultra-low field magnetic resonance imaging (ulf-mri)
US7796028B1 (en) 2006-08-01 2010-09-14 Battelle Energy Alliance, Llc Circuitry, systems and methods for detecting magnetic fields
WO2008022410A1 (en) * 2006-08-25 2008-02-28 Commonwealth Scientific And Industrial Research Organisation Method and apparatus for nano-scale squid
US7615385B2 (en) 2006-09-20 2009-11-10 Hypres, Inc Double-masking technique for increasing fabrication yield in superconducting electronics
US7629586B2 (en) 2006-11-10 2009-12-08 Gamma Medica-Ideas, Inc. Methods and systems of combining magnetic resonance and nuclear imaging
WO2008061275A1 (en) 2006-11-20 2008-05-29 Technological Resources Pty Limited A gravity gradiometer
CA2690958C (en) * 2006-11-20 2015-06-30 Technological Resources Pty Limited A gravity gradiometer
WO2008061273A1 (en) * 2006-11-20 2008-05-29 Technological Resources Pty Limited A gravity gradiometer
CA2691032A1 (en) 2006-11-20 2008-05-29 Technological Resources Pty Limited A gravity gradiometer
CA2690893A1 (en) * 2006-11-20 2008-05-29 Technological Resources Pty Limited A gravity gradiometer
CA2690886A1 (en) * 2006-11-20 2008-05-29 Technological Resources Pty Limited A gravity gradiometer
WO2008061279A1 (en) * 2006-11-23 2008-05-29 Technological Resources Pty. Limited Gravity gradiometer
EP2092372B1 (en) 2006-11-23 2012-10-24 Technological Resources PTY. Limited Gravity gradiometer
JP2010510496A (en) * 2006-11-23 2010-04-02 テクノロジカル リソーシーズ プロプライエタリー リミテッド Gravity gradiometer
ZA200710804B (en) * 2006-11-23 2009-09-30 Tech Resources Pty Ltd Compensation for unwanted accelerations in a gravity gradiometer
EP2089741B1 (en) 2006-11-23 2015-03-25 Technological Resources PTY. Limited Gravity gradiometer
EP2084560A4 (en) 2006-11-23 2010-12-15 Tech Resources Pty Ltd Gravity gradiometer
WO2008061284A1 (en) * 2006-11-23 2008-05-29 Technological Resources Pty Limited Gravity gradiometer
US8633693B2 (en) * 2007-04-02 2014-01-21 The Regents Of The University Of California Rotating-frame gradient fields for magnetic resonance imaging and nuclear magnetic resonance in low fields
BRPI0810892A2 (en) * 2007-05-02 2014-10-21 Derek D Feng Method and apparatus for continuous precision NMR / MRI based on quantum theory
WO2008137485A2 (en) * 2007-05-04 2008-11-13 California Institute Of Technology Low field squid mri devices, components and methods
WO2009023303A2 (en) * 2007-05-04 2009-02-19 California Institute Of Technology Geometries for superconducting sensing coils for squid-based systems
US7688069B2 (en) * 2007-05-18 2010-03-30 Los Alamos National Security, Llc Ultra-low field nuclear magnetic resonance and magnetic resonance imaging to discriminate and identify materials
US7994786B2 (en) * 2007-06-19 2011-08-09 Mary Hitchcock Memorial Hospital System and method for use of nanoparticles in imaging and temperature measurement
US20100321017A1 (en) * 2007-08-31 2010-12-23 Alexander Pines Ultrahigh time resolution magnetic resonance
AU2008305096B2 (en) * 2007-09-25 2011-06-09 Japan Oil, Gas And Metals National Corporation Electromagnetic prospecting apparatus and electromagnetic prospecting method
US8547090B2 (en) * 2007-12-03 2013-10-01 President And Fellows Of Harvard College Electronic spin based enhancement of magnetometer sensitivity
GB0810322D0 (en) * 2008-06-05 2008-07-09 Cryogenic Ltd Magnetic resonance imaging apparatus and method
US8138760B2 (en) * 2008-06-23 2012-03-20 Northrop Grumman Guidance And Electronics Company, Inc. Temperature system with magnetic field suppression
US9167983B2 (en) * 2008-08-15 2015-10-27 The University Of Houston System Imaging method for obtaining spatial distribution of nanoparticles in the body
US8179133B1 (en) 2008-08-18 2012-05-15 Hypres, Inc. High linearity superconducting radio frequency magnetic field detector
US8305078B2 (en) * 2008-10-09 2012-11-06 Los Alamos National Security, Llc Method of performing MRI with an atomic magnetometer
US9140657B2 (en) * 2009-04-13 2015-09-22 The Regents Of The University Of California Detection of J-coupling using atomic magnetometer
US8222899B2 (en) * 2009-04-21 2012-07-17 Herng-Er Horng Squid detected nuclear magnetic resonance and imaging at ultra-weak fields
US20100292564A1 (en) * 2009-05-18 2010-11-18 Cantillon Murphy Padraig J System and Method For Magnetic-Nanoparticle, Hyperthermia Cancer Therapy
EP2476005B1 (en) * 2009-09-09 2013-06-26 Forschungszentrum Jülich GmbH Squid with a coil inductively coupled to the squid via a mutual inductance
KR101050153B1 (en) * 2009-09-21 2011-07-19 한국표준과학연구원 Magnetic field generating device, magnetic field generating method, low magnetic field magnetic resonance device, and low magnetic field magnetic resonance imaging device
US8593141B1 (en) 2009-11-24 2013-11-26 Hypres, Inc. Magnetic resonance system and method employing a digital squid
AU2010327357B2 (en) * 2009-12-02 2014-09-18 Commonwealth Scientific And Industrial Research Organisation An apparatus for detecting signals
US8970217B1 (en) 2010-04-14 2015-03-03 Hypres, Inc. System and method for noise reduction in magnetic resonance imaging
US8022703B1 (en) * 2010-05-06 2011-09-20 Kai-Wen Huang Method for rapid detecting tumor
CN102068256B (en) * 2011-01-31 2012-10-24 汕头市超声仪器研究所有限公司 Handheld ultralow field MRI (magnetic resonance imaging) system based on SQUID (superconducting quantum interference device)
DE102011005888B4 (en) * 2011-03-22 2014-01-09 Bruker Biospin Ag Cooling of a Cryo Probe Head in a Magnetic Resonance Resonance Equipment
JP2012211813A (en) * 2011-03-31 2012-11-01 Toyohashi Univ Of Technology Squid detection nuclear magnetic resonance apparatus in super low magnetic field
CN102253417B (en) * 2011-04-23 2013-03-06 汕头市超声仪器研究所有限公司 Security checking method based on handheld ultra low field magnetic resonance imaging (MRI) system
US8493066B2 (en) 2011-06-06 2013-07-23 The United States Of America As Represented By The Secretary Of The Navy Magnetic field detection using faraday effect
US8674688B2 (en) 2011-06-06 2014-03-18 The United States Of America As Represented By The Secretary Of The Navy Measurement of small wavelength difference in coherent light using faraday effect
US9254097B2 (en) 2011-09-19 2016-02-09 Los Alamos National Security, Llc System and method for magnetic current density imaging at ultra low magnetic fields
US8749236B2 (en) * 2011-10-06 2014-06-10 Kabushiki Kaisha Toshiba MRI with fat suppression using fat decoupling RF during pre-sequence shimming
KR101287426B1 (en) * 2011-10-26 2013-07-18 한국표준과학연구원 Target Discrimination Method by using Ultra-Low Field NMR and Ultra-Low Field NMR Target Discrimination Apparatus
FI125397B (en) 2012-01-24 2015-09-30 Elekta Ab Method for using spatial and temporal oversampling in multichannel measurements
KR101310750B1 (en) * 2012-01-31 2013-09-24 한국표준과학연구원 biomagnetic resonance apparatus and the measuring method of the same
US9557394B2 (en) 2012-04-25 2017-01-31 U.S. Department Of Energy Classification of materials using nuclear magnetic resonance dispersion and/or x-ray absorption
US9411031B2 (en) 2012-04-25 2016-08-09 Los Alamos National Security, Llc Hypothesis-driven classification of materials using nuclear magnetic resonance relaxometry
CA2879008A1 (en) * 2012-07-11 2014-01-16 Nativis, Inc. Miniaturized molecular interrogation and data system
US9329246B2 (en) * 2012-10-03 2016-05-03 Bruker Biospin Ag Method for hyperpolarization transfer in the liquid state
WO2014099463A1 (en) * 2012-12-18 2014-06-26 The General Hospital Corporation System and method for sensitivity-enhanced multi-echo chemical exchange saturation transfer (mecest) magentic resonance imaging
CN108310655B (en) 2013-03-15 2021-11-26 纳特维斯公司 Controller and flexible coil for performing therapy, such as cancer therapy
CN104345195B (en) * 2013-08-09 2018-01-30 中国科学院上海微系统与信息技术研究所 Reducible incremental magnetic field cancellation element
US20170000377A1 (en) * 2014-03-14 2017-01-05 The General Hospital Corporation System and method for imaging free radicals
JP6791760B2 (en) * 2014-03-14 2020-11-25 ザ ジェネラル ホスピタル コーポレイション Systems and methods for low-field multi-channel imaging
CN105137374B (en) * 2014-06-03 2018-09-25 中国科学院上海微系统与信息技术研究所 A kind of MR imaging method and device of ultrahigh resolution
WO2016037101A1 (en) 2014-09-05 2016-03-10 Hyperfine Research, Inc. Thermal management methods and apparatus
WO2016077417A1 (en) 2014-11-11 2016-05-19 Hyperfine Research, Inc. Low field magnetic resonance methods and apparatus
CN105093093B (en) * 2015-07-17 2018-01-30 中国科学院上海微系统与信息技术研究所 A kind of SQUID chips and its detection method
US10585153B2 (en) 2016-11-22 2020-03-10 Hyperfine Research, Inc. Rotatable magnet methods and apparatus for a magnetic resonance imaging system
US10539637B2 (en) 2016-11-22 2020-01-21 Hyperfine Research, Inc. Portable magnetic resonance imaging methods and apparatus
US10627464B2 (en) 2016-11-22 2020-04-21 Hyperfine Research, Inc. Low-field magnetic resonance imaging methods and apparatus
CN106873044B (en) * 2017-04-19 2019-06-11 吉林大学 Array SQUID nuclear magnetic resonance underground water detecting device and imaging method
US10973435B2 (en) 2017-04-28 2021-04-13 The Penn State Research Foundation Adult head-sized coil-based low-field MRI
US11723579B2 (en) 2017-09-19 2023-08-15 Neuroenhancement Lab, LLC Method and apparatus for neuroenhancement
US11717686B2 (en) 2017-12-04 2023-08-08 Neuroenhancement Lab, LLC Method and apparatus for neuroenhancement to facilitate learning and performance
US11273283B2 (en) 2017-12-31 2022-03-15 Neuroenhancement Lab, LLC Method and apparatus for neuroenhancement to enhance emotional response
US11364361B2 (en) 2018-04-20 2022-06-21 Neuroenhancement Lab, LLC System and method for inducing sleep by transplanting mental states
KR20210013563A (en) 2018-05-21 2021-02-04 하이퍼파인 리서치, 인크. B0 magnet methods and devices for magnetic resonance imaging system
WO2020056418A1 (en) 2018-09-14 2020-03-19 Neuroenhancement Lab, LLC System and method of improving sleep
CN109864733A (en) * 2019-01-16 2019-06-11 漫迪医疗仪器(上海)有限公司 Detection method, system, medium and the equipment of heart and brain exception
US11786694B2 (en) 2019-05-24 2023-10-17 NeuroLight, Inc. Device, method, and app for facilitating sleep
US11093794B1 (en) * 2020-02-13 2021-08-17 United States Of America As Represented By The Secretary Of The Navy Noise-driven coupled dynamic pattern recognition device for low power applications
FR3117218B1 (en) 2020-12-03 2023-03-17 Chipiron SQUID-based low-noise RF detection and acquisition system and equipment integrating this system
DE102021118736A1 (en) 2021-07-20 2023-01-26 Leibniz-Institut für Angewandte Geophysik Method and system for determining a soil water content by means of NMR
CN113768505A (en) * 2021-08-11 2021-12-10 季华实验室 Wearable magnetoencephalography environmental noise suppression system and method

Family Cites Families (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR1475608A (en) * 1964-04-30 1967-04-07 Inst Francais Du Petrole Hydrocarbon detection method
SU742839A1 (en) 1978-04-11 1980-06-25 Научно-Производственное Объединение "Геофизика" Method of measuring magnetic fields using nuclear magnetic resonance
US4442404A (en) * 1978-12-19 1984-04-10 Bergmann Wilfried H Method and means for the noninvasive, local, in-vivo examination of endogeneous tissue, organs, bones, nerves and circulating blood on account of spin-echo techniques
DE2951537A1 (en) * 1979-12-20 1981-07-02 Siemens AG, 1000 Berlin und 8000 München TEMPERATURE METHOD
GB2128745B (en) * 1982-08-31 1986-09-17 Asahikawa Medical College Method of measuring internal information from a body by using nuclear magnetic resonance
US4588947A (en) * 1983-12-30 1986-05-13 International Business Machines Corporation Integrated miniature DC SQUID susceptometer for measuring properties of very small samples
DE3529815A1 (en) * 1985-08-20 1987-02-26 Siemens Ag MEASURING DEVICE WITH A SQUID MAGNETOMETER
US4851777A (en) * 1986-04-21 1989-07-25 Stanford University Reduced noise NMR localization system
FI874419A (en) * 1987-10-08 1989-04-09 Instrumentarium Oy ANORDNING OCH FOERFARANDE FOER UNDERSOEKNING AV ETT OBJEKTS EGENSKAPER.
US4987368A (en) * 1987-11-05 1991-01-22 Shell Oil Company Nuclear magnetism logging tool using high-temperature superconducting squid detectors
FR2621392A1 (en) * 1988-09-30 1989-04-07 Gen Electric Cgr LOW FIELD NMR MACHINE WITH DYNAMIC POLARIZATION
GB2227095B (en) 1989-01-03 1993-01-27 Instrumentarium Corp Magnetic resonance imaging apparatus and method
US5057776A (en) * 1989-12-08 1991-10-15 Albert Macovski Pulsed field MRI system with non-resonant excitation
US5773024A (en) 1989-12-22 1998-06-30 Imarx Pharmaceutical Corp. Container with multi-phase composition for use in diagnostic and therapeutic applications
US5122744A (en) 1990-10-09 1992-06-16 Ibm Corporation Gradiometer having a magnetometer which cancels background magnetic field from other magnetometers
US5254950A (en) * 1991-09-13 1993-10-19 The Regents, University Of California DC superconducting quantum interference device usable in nuclear quadrupole resonance and zero field nuclear magnetic spectrometers
US5300887A (en) * 1992-04-27 1994-04-05 Albert Macovski Pulsed field MRI system with spatial selection
US5343147A (en) * 1992-09-08 1994-08-30 Quantum Magnetics, Inc. Method and apparatus for using stochastic excitation and a superconducting quantum interference device (SAUID) to perform wideband frequency response measurements
WO1995003550A1 (en) * 1993-07-22 1995-02-02 Josef Constantin Szeles Nuclear magnetic resonance process and device for imaging the body
US5600243A (en) * 1993-09-07 1997-02-04 Conductus, Inc. Magnetically shielded magnetic sensor with squid and ground plane
US5557199A (en) * 1994-04-29 1996-09-17 The United States Of America As Represented By The Department Of Health And Human Services Magnetic resonance monitor
AT406092B (en) * 1995-08-28 2000-02-25 Szeles Josef Constantin Dr METHOD AND DEVICE FOR FORMING AN IMAGE WITH NUCLEAR SPIN RESONANCE
US6426058B1 (en) 1996-03-29 2002-07-30 The Regents Of The University Of California Enhancement of NMR and MRI in the presence of hyperpolarized noble gases
GB2351549B (en) 1996-08-15 2001-02-14 Univ Aberdeen Liquified gas cryostat
GB9617175D0 (en) 1996-08-15 1996-09-25 Univ Aberdeen Liquid gas cryostat
JP2993444B2 (en) * 1996-10-25 1999-12-20 日本電気株式会社 Connection setting and recovery method in ATM network
US5835995A (en) * 1996-10-28 1998-11-10 Macovski; Albert Localized pulsed superconductive MRI system
US6023161A (en) * 1997-02-28 2000-02-08 The Regents Of The University Of California Low-noise SQUID
US6159444A (en) * 1998-09-11 2000-12-12 The Regents Of The University Of California NMR/MRI with hyperpolarized gas and high Tc SQUID
US6284222B1 (en) 1998-11-03 2001-09-04 Medi--Physics, Inc. Hyperpolarized helium-3 microbubble gas entrapment methods
JP4068763B2 (en) * 1999-06-21 2008-03-26 株式会社島津製作所 Biological signal measuring device
JP4193382B2 (en) * 2001-07-19 2008-12-10 株式会社日立製作所 Magnetic field measuring device
EA006832B1 (en) * 2002-02-06 2006-04-28 Те Риджентс Оф Те Юниверсити Оф Калифорния Squid detected nmr and mri at ultraflow fields
US6724188B2 (en) * 2002-03-29 2004-04-20 Wavbank, Inc. Apparatus and method for measuring molecular electromagnetic signals with a squid device and stochastic resonance to measure low-threshold signals

Also Published As

Publication number Publication date
US20070018643A1 (en) 2007-01-25
WO2003067267A3 (en) 2003-11-06
ZA200406984B (en) 2005-09-20
CN1643403A (en) 2005-07-20
NZ534963A (en) 2006-02-24
US7466132B2 (en) 2008-12-16
US20050134262A1 (en) 2005-06-23
EP1474707A2 (en) 2004-11-10
AU2003210906A1 (en) 2003-09-02
AU2003210906B2 (en) 2010-06-03
US20040027125A1 (en) 2004-02-12
US6885192B2 (en) 2005-04-26
AP2004003127A0 (en) 2004-09-30
IL163669A0 (en) 2005-12-18
US7116102B2 (en) 2006-10-03
US7053610B2 (en) 2006-05-30
US20060176054A1 (en) 2006-08-10
CA2478466A1 (en) 2003-08-14
EP1474707B1 (en) 2012-07-18
EA006832B1 (en) 2006-04-28
US20080074113A1 (en) 2008-03-27
MXPA04008593A (en) 2004-12-06
US7218104B2 (en) 2007-05-15
WO2003067267A2 (en) 2003-08-14
EP1474707A4 (en) 2005-05-25
EA200401028A1 (en) 2005-04-28
CN1287160C (en) 2006-11-29

Similar Documents

Publication Publication Date Title
CA2478466C (en) Squid detected nmr and mri at ultralow fields
US7187169B2 (en) NMR and MRI apparatus and method
McDermott et al. Microtesla MRI with a superconducting quantum interference device
McDermott et al. SQUID-detected magnetic resonance imaging in microtesla magnetic fields
Matlachov et al. SQUID detected NMR in microtesla magnetic fields
US8222899B2 (en) Squid detected nuclear magnetic resonance and imaging at ultra-weak fields
US6159444A (en) NMR/MRI with hyperpolarized gas and high Tc SQUID
EP2294437B1 (en) Magnetic resonance imaging apparatus and method using squid detection and field- cycling
Augustine et al. Low field magnetic resonance images of polarized noble gases obtained with a dc superconducting quantum interference device
Augustine et al. Squid detected nmr and nqr
Wong-Foy et al. Laser-polarized 129Xe NMR and MRI at ultralow magnetic fields
WO2006052236A1 (en) Nmr and mri apparatus and method involving a squid magnetometer
De Souza et al. NMR and MRI obtained with high transition temperature dc SQUIDs
Schlenga et al. High-T/sub c/SQUIDs for low-field NMR and MRI of room temperature samples
Lusher et al. Broadband nuclear magnetic resonance using DC SQUID amplifiers
EP4279938A1 (en) Apparatus and method for investigating a sample by using nuclear magnetic resonance and a magnetoresistance sensor
McDermott SQUID-detected NMR and MRI in microtesla magnetic fields
Seifert et al. Parallel Imaging with SQUID Arrays
Frolov SPECTROMETER FOR NUCLEAR MAGNETIC RESONANCE IN INTERMEDIATE FIELDS
McDermott et al. Microtesla magnetic resonance imaging with a superconducting quantum interference device
Weisskoff et al. Rf SQUID (superconducting quantum interference device) detector for single-ion trapping experiments
Kuznetsov Quantum magnetometer for measuring magnetic susceptibility over wide ranges of temperatures and fields
Seki SQUID Applications-Ultra-low-Field NMR Measurements of Protons and Fluorine Nuclei using SQUID

Legal Events

Date Code Title Description
EEER Examination request
MKLA Lapsed

Effective date: 20160208