Prepolarized magnetic resonance imaging (PMRI) uses two pulsed electromagnets to achieve high-field image quality with the benefits of low-field data acquisition. The principal challenge with all resistive MRI systems is the implementation of a highly precise magnet current supply. The noise current through the magnet is fundamentally limited by the current transducer used to provide feedback and the voltage reference used to generate the demand signal. Field instability in the main field magnet can both corrupt the received data and degrade the robustness of Carr¿Purcell¿Meiboom¿Gill (CPMG) echo trains, which are paramount to efficient imaging in PMRI. In this work, we present the magnet control system that achieved sufficient field stability for PMRI at $0.5/0.13$ T, identify the dominant sources of noise in the control system, examine the imaging artifacts that can occur if the field stability is insufficient, and identify how the design can be improved for better field stability, should it be required for future implementations of PMRI.

We describe the electronics for controlling the independently pulsed polarizing coil in a prepolarized magnetic resonance imaging (PMRI) system and demonstrate performance with free induction decay measurements and in vivo imaging experiments. A PMRI scanner retains all the benefits of acquiring MRI data at low field, but with the higher signal of the polarizing field. Rapidly and efficiently ramping the polarizing coil without disturbing the data acquisition is one of the major challenges of PMRI. With our modular hardware design, we successfully ramp the 0.4-T polarizing coil of a wrist-sized PMRI scanner at up to 100 T/s without causing image artifacts or otherwise degrading data acquisition.

A graphical user interface (GUI) has been developed which enables interactive feedback and control to the real-time diminishing variance algorithm (DVA). This interactivity allows the user to set scan parameters, view scan statistics, and view image updates during the course of the scan. In addition, the DVA has been extended to simultaneously reduce motion artifacts in three dimensions using three orthogonal navigators. Preliminary in vivo studies indicate that these improvements to the standard DVA allow for significantly improved consistency and robustness in eliminating respiratory motion artifacts from MR images, particularly when imaging the coronary arteries.

Methods for improving the contrast-to-noise ratio (CNR) of low-contrast lesions in medical ultrasound imaging are described. Differences in the frequency spectra and amplitude distributions of the lesion and its surroundings can be used to increase the CNR of the lesion relative to the background. Automated graylevel mapping is used in combination with a contrast-weighted form of frequency-diversity speckle reduction. In clinical studies, the techniques have yielded mean CNR improvements of 3.2 dB above ordinary frequency-diversity imaging and 5.6 dB over sharper conventional images, with no post-processing graylevel mapping.

A spectral extrapolation algorithm for spatially bounded images is presented. An image is said to be spatially bounded when it is confined to a closed region and is surrounded by a background of zeros. With prior knowledge of the spatial domain zeros, the extrapolation algorithm extends the image's spectrum beyond a known interval of low-frequency components. The result, which is referred to as the finite support solution, has space variant resolution; features near the edge of the support region are better resolved than those in the center. The resolution of the finite support solution is discussed as a function of the number of known spatial zeros and known spectral components. A regularized version of the finite support solution is included for handling the case where the known spectral components are noisy. For both the noiseless and noisy cases, the resolution of the finite support solution is measured in terms of its impulse response characteristics, and compared to the resolution of the zerofilled and Nyquist solutions. The finite support solution is superior to the zerofilled solution for both the noisy and noiseless data cases. When compared to the Nyquist solution, the finite support solution may be preferred in the noisy data case. Examples using medical image data are provided.

While acquiring data for magnetic resonance spectroscopic (MRS) images, signal-to-noise (SNR) and imaging time considerations limit the amount of sampled spatial frequencies. Typically, the small number of spatial frequencies are uniformly sampled in the center of the transform domain (k-space) and the resulting image, generated from zero-filling, appears blurry. Often times, the MRS image is further degraded by large interfering components. To improve these images, the authors use knowledge of the anticipated anatomical features in the spectroscopic image, as determined from a conventional magnetic resonance image (MRI). They use this prior knowledge during both data acquisition and reconstruction. They show results where the present method not only improves the resolution of the MRS image, but also reduces contamination from undesirable components, such as subcutaneous lipid.>

An estimation approach is described for three-dimensional reconstruction from line integral projections using incomplete and very noisy data. Generalized cylinders parameterized by stochastic dynamic models are used to represent prior knowledge about the properties of objects of interest in the probed domain. The object models, a statistical measurement model, and the maximum a posteriori probability performance criterion are combined to reformulate the reconstruction problem as a computationally challenging nonlinear estimation problem. For computational feasibility, a suboptimal hierarchical algorithm is described whose individual steps are locally optimal and are combined to satisfy a global optimality criterion. The formulation and algorithm are restricted to objects whose center axis is a single-valued function of a fixed spatial coordinate. Simulation examples demonstrate accurate reconstructions with as few as four views in a 135 degrees sector, at an average signal-to-noise ratio of 3.3.>

A novel magnetic resonance imaging system is explored which uses data collected in the presence of quadrature phase sinusoidal gradients along an excited plane. It is shown that the resulting magnetization signal (FID) is a sum of the Hankel transforms of the radial modulators of the imaged plane. An efficient processing system is suggested which extracts the different Hankel transforms from the received signal. These transforms are used to directly reconstructs the two dimensional (2-D) image using either 1-D inverse Hankel transforms or a 2-D inverse Fourier transform. The main advantages of the proposed system are the potential for rapid imaging (not discussed here), the ability to trade-off acquisition time and SNR and resonant gradient circuits.

A unified framework for the exact Maximum Likelihood estimation of the parameters of superimposed exponential signals in noise, encompassing both the single and the multiexperiment cases (respectively the time series and the array problems), is presented. An exact expression for the ML criterion is derived in terms of the prediction polynomial of the noiseless signal, and an iterative algorithm for the maximization of this criterion is presented. A simulation example shows the estimator to be capable of providing more accurate frequency estimates than currently existing techniques.

An estimation approach to three dimensional reconstruction from projections, with incomplete and very noisy data, is suggested. Using a stochastic dynamic model for the object of interest in the probed domain, the reconstruction problem is reformulated as a nonlinear state estimation problem of small dimensionality, and an approximate MMSE globally optimal algorithm for its solution is presented. The algorithm, which is recursive in a hybrid frequency-space domain, operates directly on the Fourier transformed projection data, eliminating altogether the attempt to invert the projection integral equation. The computational requirements compare favorably with those of conventional reconstruction procedures, which fail in the limited and noisy data case.

An operator framework for image reconstruction from limited line integral data is presented in this paper. From this framework we derive iterative limited data reconstruction algorithms. The framework applies to arbitrary scanning geometries and arbitrary patterns of missing data. All available a priori information is incorporated by constraint operators. We show how a priori knowledge of the density in part of the object improves the quality of the reconstructed image. The technique is demonstrated with real x-ray data collected on a clinical CT scanner.

Spatial frequency encoding is described as it applies to both photographic and television systems. In each case, color separations amplitude modulate spatial frequency carriers. The photographic system separates the carriers in the two-dimensional Fourier transform plane using coherent optics; the television system employs a two-dimensional electrical filter to perform the same separation. These filters are synthesized so that the two-dimensional bandwidths of the television camera are used efficiently. Nonlinear processing techniques are also described to minimize crosstalk between channels.