The observer SNR penalty for reconstructions from projections

K-space filter deconvolution and flip angle self-calibration in 2D radial hyperpolarised 3He lung MRI

In hyperpolarised (3)He lung MRI with constant flip angles, the transverse magnetisation decays with each RF excitation imposing a k-space filter on the acquired data. For radial data acquired in an angularly-sequential order, this filter causes streaking, angular shading and loss of spatial resolution in the images. The main aim of this work was to reduce the effects of the RF depletion k-space filter in radial acquisitions. Two approaches are presented; (i) retrospective deconvolution of the k-space filter for sequentially-acquired data and (ii) golden angle acquisition order. Radial trajectories sample the centre of k-space with every projection, thereby self-tracking signal decay. The inverse of the signal decay function was used to retrospectively deconvolve RF depolarisation k-space filter effects and the method was demonstrated in 2D radial imaging in phantoms and human lungs. A golden angle radial acquisition was shown to effectively suppress artefacts caused by the RF depletion k-space filter. In addition, the average flip angle per slice was calculated from the signal decay and the values were found to correspond with conventional flip angle maps, providing a means of flip angle self-calibration.

Rapid Fourier imaging using steady-state free precession

Reversal of the read gradient in a SSFP imaging experiment allows a full spin echo to be collected in the interval tau between successive rf pulses. Orthogonal gradient pulses are used to dephase and subsequently rephase the transverse magnetization each tau enabling 2D or 3D Fourier techniques. The minimum data collection time per slice in the 3D technique is 3.1 s (128 X 256). For a 2D data collection, an oscillating bipolar sawtooth gradient is used to select the slice. Each phase-encode value must be averaged over an equivalent portion of the oscillating slice-selection gradient and this condition gives a minimum of 25 s for 2D data collection. Excellent slice selection is achieved with less than 5% of the signal lying outside the slice profile central lobe. Images at 0.14 T show tissue contrast may be manipulated by changing the rf pulse angle, an example of which is the presence or absence of gray/white matter contrast at rf pulse angles of 30 and 90 degrees, respectively. The pulse angle theta dependence of five samples with different values of T2/T1 was measured and numerically calculated with good agreement between theory and experiment for theta less than or equal to 90 degrees.

Noise and filtration in magnetic resonance imaging

Noise in two-dimensional Fourier transform magnetic resonance images has been investigated using noise power spectra and measurements of standard deviation. The measured effects of averaging, spatial filtering, temporal filtering, and sampling have been compared with theoretical calculations. The noise of unfiltered images is found to be white, as expected, and the choice of the temporal filter and sampling interval affects the noise in a manner predicted by sampling theory. The shapes of the imager's spatial frequency filters are extracted using noise power spectra.

Chemical-shift imaging using filtered back-projection algorithms

A method of performing chemical-shift imaging is described that avoids some of the disadvantages inherent in the techniques which use spin echos and phase encoding. A series of FIDs are recorded during the application of a series of static gradients and the spectral image is reconstructed using a conventional filtered back-projection algorithm.

Unified snr analysis of medical imaging systems

The ideal observer signal to noise ratio (snr) has been derived from statistical decision theory for all of the major medical imaging modalities. This snr provides an absolute scale for image system performance assessment and leads to instrumentation design goals and constraints for imaging system optimisation since no observer can surpass the performance of the ideal observer. The dependence of detectable detail size on exposure or imaging time follows immediately from the analysis. A framework emerges for comparing data acquisition techniques, e.g. reconstruction from projections versus Fourier methods in nmr imaging, and time of flight positron emission tomography (tofpet) versus conventional pet. The approach of studying the ideal observer is motivated by measurements on human observers which show that they can come close to the performance of the idea) observer, except when the image noise has negative correlations-as in images reconstructed from projections-where they suffer a small but significant penalty.