Accelerating hyperpolarized metabolic imaging of the heart by exploiting spatiotemporal correlations

Hyperpolarized (13)C-labeled pyruvate is a promising tool to investigate cardiac metabolism. It has been shown that changes in substrate metabolism occur following the induction of ischemia. To investigate the metabolic changes that are confined to spatial regions, high spatiotemporal resolution is required. The present work exploits both spatial and temporal correlations using k-t principal component analysis (PCA) to undersample the spatiotemporal domain, thereby speeding up data acquisition. A numerical model was implemented to investigate optimal acquisition and reconstruction parameters for pyruvate, lactate and bicarbonate maps of the heart. Subsequently, prospectively undersampled in vivo data on rat hearts were acquired using a combination of spectral-spatial signal excitation and a variable-density single-shot echo planar readout. Using five-fold k-t PCA, a spatial resolution of 1 × 1 mm(2) at a temporal resolution of 3 s was achieved.