Search

Dynamic contrast-enhanced (DCE) MR cardiac imaging has been recognized as a unique and powerful tool for assessing both cardiac functions and physiological conditions of the heart tissues (e.g., tissue rejection following heart transplantation). However, because of cardiac motion and the limited data acquisition speed of existing MRI techniques, it has been very difficult to acquire dynamic images of very high spatiotemporal resolution. This paper proposes a new generalized series (GS) based imaging technique to overcome this challenging problem. Specifically, the proposed technique collects two data sets: a) a sequence of highresolution reference images over several cardiac cycles using a gated cine acquisition scheme before the injection of a contrast agent (or a molecular probe), and b) a sequence of reduced data sets with very high frame rate during the transient wash-in/wash-out stage of the contrast agent. A GS model is then used to combine these two data sets to reconstruct a high-resolution image sequence, capturing both the cardiac motions and dynamic signal changes due to the interaction of the contrast agent with the cardiac tissues. The proposed technique has been validated using both simulated and experimental data, which show that high-resolution dynamic images can be acquired with as few as 8 encodings (in contrast to 256 encodings required in the traditional Fourier transform-based methods). The technique provides a very effective tool for physiological imaging of the beating heart with molecular probes.