Meeting Banner
Abstract #3939

Laplace Inversion for Kinetic Analysis of Hyperpolarized 13C Data Without a Priori Knowledge Using a Hybrid Maximum Entropy Method (MEM)/ Non-Linear Least Square (NLS)

Erika Mariotti1, Fiona Shaughnessy1, Rodolfo A. Medina1, Joel T. Dunn1, Richard Southworth1, Thomas R. Eykyn1, 2

1Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom; 2CRUK and EPSRC Cancer Imaging Centre, Royal Marsden NHS Trust, The Institute of Cancer Research, Sutton, Surrey, United Kingdom

To obtain information on the underlying metabolic activities, hyperpolarized data are usually fit using kinetic models to derive rates. In this work, we show the feasibility of analysing hyperpolarized data without prior assumptions of a model and, at the same time, overcoming the multiple local minimum problem by applying a hybrid Maximum-Entropy/NonLinear-Least-Squares (MEM/NLS) method4 to hyperpolarized 13C data both in vitro in whole blood and ex vivo in the isolated rat heart.

Keywords

acid activities actual additionally agreement alanine applying assume assumptions avoids biases bicarbonate biological biology biomedical blood brain build cancer cells challenged characterize characterizing collected college compartment compartments complex component components conditioned constant containing curve curves dashed datasets decay derive derived describe describing detected determined division engineering entire entropy example excitation experimental exponential extra fact fail feasibility fitted fitting flow gadolinium good heart heterogeneous hybrid in vivo injected input institute instrument inversion isolated iteratively journal kinetic king kingdom knowledge lactate least likely linear liver local male metabolic metabolism metabolites mixed model models natl nature negative nonlinear novel organs output overcoming oxford pathway pathways peaks perfused positive possibly previously prior priori problem proposed radical rats reactions representative respectively royal sciences series simple solid solution sought spectra spectrum square squares structure suggested suggests summed summing surrey systems traditional trust underlying useful usually variety vitro whereas whole