SQUID-detected magnetic resonance imaging in microtesla magnetic fields

2004 
We describe studies of nuclear magnetic resonance (NMR) spectroscopy and magnetic resonance imaging (MRI) of liquid samples at room temperature in microtesla magnetic fields. The nuclear spins are prepolarized in a strong transient field. The magnetic signals generated by the precessing spins, which range in frequency from tens of Hz to several kHz, are detected by a low-transition temperature dc SQUID (Superconducting QUantum Interference Device) coupled to an untuned, superconducting flux transformer configured as an axial gradiometer. The combination of prepolarization and frequency-independent detector sensitivity results in a high signal-to-noise ratio and high spectral resolution (∼1 Hz) even in grossly inhomogeneous magnetic fields. In the NMR experiments, the high spectral resolution enables us to detect the 10-Hz splitting of the spectrum of protons due to their scalar coupling to a 31P nucleus. Furthermore, the broadband detection scheme combined with a non-resonant field-reversal spin echo allows the simultaneous observation of signals from protons and 31P nuclei, even though their NMR resonance frequencies differ by a factor of 2.5. We extend our methodology to MRI in microtesla fields, where the high spectral resolution translates into high spatial resolution. We demonstrate two-dimensional images of a mineral oil phantom and slices of peppers, with a spatial resolution of about 1 mm. We also image an intact pepper using slice selection, again with 1-mm resolution. In further experiments we demonstrate T1-contrast imaging of a water phantom, some parts of which were doped with a paramagnetic salt to reduce the longitudinal relaxation time T1. Possible applications of this MRI technique include screening for tumors and integration with existing multichannel SQUID systems for brain imaging.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    18
    References
    70
    Citations
    NaN
    KQI
    []