Abstract
The advent of fast sequences based on the Rapid Acquisition with Relaxation Enhancement (RARE) sequence allowed T2-weighted imaging to be performed in clinically feasible scan times and these sequences have become the workhorse for spin-echo imaging in Magnetic Resonance Imaging (MRI). In this work, we propose several variants and applications of the 3D Turbo Spin Echo (TSE) sequence designed to efficiently image endogenous and exogenous long- T2 species in the body.
There has been renewed interest in developing and utilizing non-contrast MRA methods because of the risk of nephrogenic systemic fibrosis (NSF) in patients with kidney dysfunction. TSE sequences are capable of generating excellent contrast between blood and surrounding tissue based on T2 differences, but conventional TSE sequences are susceptible to flow-related artifacts and signal loss.
Here, a new, hybrid pulse sequence that shares characteristics with the TSE and balanced Steady State Free Precession (bSSFP) sequence is introduced. Desirable T2 contrast is provided by the TSE-like pulse sequence design and the improved flow-performance characteristics of bSSFP imaging are gained through fully-refocused gradients and refocusing pulse phase alternation of RF pulses. The principles of flow-independent angiography in the context of this new TSE sequence, including a description of tissue contrast properties, are described. Additionally, a method for fat/water separation based on phase detection is presented for this application which relies on catalyzation of the TSE echo train. This sequence is capable of performing non-contrast enhanced angiography of the peripheral arterial tree by exploiting the T2 differences between blood and surrounding tissue.
In order to take advantage of some of the unique properties afforded by non-Cartesian imaging, a 3D TSE sequence which utilizes spiral readouts was developed. For peripheral angiography, spoiler gradients placed in the through-plane direction remove spurious fat signal, and longer echo spacings are available due to spiral sampling, leading to improved venous suppression. To most efficiently acquire the data, partial k-space coverage along with variable density spirals are used to undersample k-space in a controlled and well-understood manner. Additional techniques to prevent spiral artifacts (blurring) from concomitant and B0 fields as well as corrections to spiral trajectories are applied.
The major focus of MRI research has been on improving resolution, contrast, and shortening acquisition times of standard anatomical images; however, patient outcomes may also be improved by acquiring secondary, functional information about the patient. Yet functionalizing MRI is still in early stages, and much work needs to be done before it is ready to enter the clinic in an impactful way. In a departure from angiography, the TSE sequence is used as an efficient method to generate and utilize signal from fluorine, a secondary nucleus which is of interest for cellular and molecular imaging. Here, the TSE sequence will be used to efficiently sample the persistent signal arising from the relatively long T2 of fluorine. An SNR efficiency comparison between common fluorine MRI sequences is presented along with an application of model-based iterative reconstruction techniques designed to improve the quality of low-SNR fluorine images by incorporating boundary information derived from accompanying proton images.
The contributions described in this work extend the utility of TSE-type sequences in non-traditional imaging situations by exploiting some of the unique capabilities and attributes of the sequence. Furthermore, the applications described here are directly applicable to unmet clinical and pre-clinical needs.
