A.J. Lucas

Development of a Combined microPET®-MR System

As evidenced by the success of PET-CT, there are many benefits from combining imaging modalities into a single scanner. The combination of PET and MR offers potential advantages over PET-CT, including improved soft tissue contrast, access to the multiplicity of contrast mechanisms available to MR, simultaneous imaging and fast MR sequences for motion correction. In addition, PET-MR is more suitable than PET-CT for cancer screening due to the elimination of the radiation dose from CT. A key issue associated with combining PET and MR is the fact that the performance of the photomultiplier tubes (PMTs) used in conventional PET detectors is degraded in the magnetic field required for MR. Two approaches have been adopted to circumvent that issue: retention of conventional, magnetic field-sensitive PMT-based PET detectors by modification of other features of the MR or PET system, or the use of new, magnetic field-insensitive devices in the PET detectors including avalanche photo-diodes (APDs) and silicon photomultipliers (SiPMs). Taking the former approach, we are assembling a modified microPET® Focus 120 within a gap in a novel, 1T superconducting magnet. The PMTs are located in a low magnetic field (~30mT) through a combination of magnet design and the use of fiber optic ‘bundles’. Two main features of the modified PET system have been tested, namely the effect of using long fiber optic bundles in the PET detector, and the impact of magnetic field upon the performance of the position sensitive PMTs. The design of a modified microPET®-MR system for small animal imaging is completed, and assembly and testing is underway.

Split gradient coils for simultaneous PET-MRI

Combining positron emission tomography (PET) and MRI necessarily involves an engineering tradeoff as the equipment needed for the two modalities vies for the space closest to the region where the signals originate. In one recently described scanner configuration for simultaneous positron emission tomography–MRI, the positron emission tomography detection scintillating crystals reside in an 80‐mm gap between the 2 halves of a 1‐T split‐magnet cryostat. A novel set of gradient and shim coils has been specially designed for this split MRI scanner to include an 110‐mm gap from which wires are excluded so as not to interfere with positron detection. An inverse boundary element method was necessarily employed to design the three orthogonal, shielded gradient coils and shielded Z0 shim coil. The coils have been constructed and tested in the hybrid positron emission tomography‐MRI system and successfully used in simultaneous positron emission tomography‐MRI experiments. Magn Reson Med, 2009.

Pore Geometry Information Via Pulsed-Field Gradient Nmr

Studies of echo attenuation at long diffusion times in pulsed field gradient NMR experiments on a variety of rock core samples are interpreted in the light of recent theoretical analysis of the effect of pore geometry and surface relaxation. This study is motivated by the need to test the applicability of that theory to real rock systems.

Strategies for overcoming linewidth limitations in quantitative petrophysical NMR measurements.

The simple pulse-acquire experiment has been used to evaluate the level of accuracy and precision achievable in NMR fluid saturation measurements for a range of rock core samples saturated with either brine or hydrocarbons. For a set of more than 70 cores measured at 0.66 T the mean error in the NMR measurement is only 0.35% porosity when the sample linewidths are less than 50 ppm. However, for a significant portion of cores, those with very broad NMR linewidths (> 50 ppm), difficulties associated with nonuniform excitation are encountered. The magnetic susceptibility difference between pore fluid and rock matrix translates into relatively broad NMR linewidths, and this feature of petrophysical samples is the major difficulty in performing quantitative NMR experiments. Numerical simulations are used to complement the experimental results in order to develop strategies for obtaining accurate NMR results with these difficult samples.

Quantitative longitudinal fluid saturation profiles with a slice-selected CPMG sequence

A technique for obtaining quantitative longitudinal saturation/porosity profiles of rock cores which are longer than the NMR coil has been developed. The slice-selected experiment uses a prefocused pulse in conjunction with a magnetic field gradient for the localization and a CPMG sequence to sample the data. A variety of rock core samples has been studied ranging from limestones to shaly sandstones. Comparison of the relaxation decay curves obtained from these experiments and the bulk experiments show that reliable localized relaxation data are obtained.

Initial performance assessment of a combined microPET® focus-F120 and MR split magnet system

This work is an initial evaluation of the performance of a microPET Focus-Fl20 system modified to fit in the split in a 1 Tesla MR imaging system to create a rodent scanner capable of simultaneous PET and MR acquisition. The MR imager is a novel design with a split in the magnet former, cryostat and gradient coils and with sufficient self-shielding to radically reduce the stray field at the cryostat edge. This permits the Focus-F120 detector geometry to be maintained with the addition of longer fibre bundles coupling the crystals to the PMT’s, which consequently are situated in a low magnetic field region outside the cryostat. Baseline PET performance was measured with gradient and shim coils in place but without RF coils in the PET FOV. Misalignment of the crystal blocks has decreased the PET spatial resolution, from 1.7 mm to 2.2 mm at 5 mm radial distance, compared to the Focus-F120. The peak absolute sensitivity is reduced to 5.3% compared to 6.8% for a 250-750 keV energy window and 6 ns coincidence window. The noise equivalent count rate (NECR-1R) peaks at about 215 kcps at 112 MBq for a NEMA rat phantom compared to 300 kcps for the Focus-F120 at 140 MBq. Magnetic field homogeneity was assessed as +/− 6 hertz over a 10 cm long and 8 cm diameter cylindrical space and a pulsed gradient strength of 200 mT/m, with a rise time of less than 100 microseconds, demonstrated on all three axes. Separate PET and MR imaging capability has been demonstrated as well as continuous, simultaneous acquisition with no detectable interference between the two imaging systems.

Very high resolution 3D list-mode PET reconstruction using polar voxels

We present EM reconstructions using polar voxels and real data from a modified microPET Focus 120 small animal PET scanner. The reconstruction is based on a complete Monte Carlo generated system matrix that is stored in memory. Details of the implementation are given. The polar reconstructions are compared to a standard 3DFBP reconstruction method.

3d Autocorrelation for the Determination of Large-Pore Sizes

A data analysis methodology is used to process 3D NMR image data acquired for porous systems. The method extracts the mean size of those repeating elements in the image data which are largely compared with the image voxel dimensions. In this work we extend the two-dimensional (2D) image analysis method described by others to three spatial dimensions (3D). 3D image data were acquired at a magnetic field strength of 7 T using NMR microscopy hardware. The 3D autocorrelation function obtained from the data reveals a characteristic pore size in each dimension.