Meeting Banner
Abstract #0769

Rapid Acquisition of PET Attenuation Maps from Highly Undersampled UTE Images Using Sparse-SENSE Reconstruction

Andrew Peter Aitken1, Christoph Kolbitsch1, Tobias Schaeffter1, Claudia Prieto1

1Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom

Attenuation correction (AC) maps that accurately depict bone are required for precise PET reconstruction. Segmented AC maps can be obtained from ultrashort echo time (UTE) MR imaging. However, the clinical use of high resolution UTE in simultaneous PET/MR is limited by the long scan time (>7 minutes for the head). We show, in prospectively undersampled acquisitions, that compressed sensing combined with parallel imaging can be used to reduce the scan time to less than one minute, while retaining accurate AC-maps with respect to those produced from fully sampled data.

Keywords

accelerated acceleration according accuracy accurate achieve acquisition acquisitions actively addition agreement around arrows attenuation automatic axial become best beyond biomedical blue bone bottom boundaries camera channel classification classified clearly clinical clustering coil college combination comparable contrast coronal correct corrected correction correctly cranial decay decreases depict derived described deteriorates diameter division dual dynamic eddy encoding engineering especially exploiting field fine finite fully future generated good head heads healthy highly incorporates independently investigate investigated isotropic king kingdom leads lesions limited long maps minimizing minute misclassification misclassified needed occurs operator orbits orientations orthogonal osseous particular percentage peter pixel pixels precisely pressed produced projections prospective prospectively radial rapid recently reconstructed reconstruction reconstructions regularization required resolution resolved retrospectively sampled sciences segmentation segmented selection sense sensing sensitivities sensitivity short simulations simultaneous sinuses skin slices smoothing soft space sparse sparsity spatial structures successfully system term threshold tissue ultrashort volunteer volunteers yellow