Optimization of Diffusion Sequences Using Bootstrap Algorithms

Carolin Reischauer¹, Robert Stefan Vorburger¹, Philipp Staempfli¹, Peter Boesiger¹

The experimental optimization of diffusion sequences is crucial with respect to prospective clinical studies. But up to now it has not been possible to quantify measurement precision of diffusion parameters in-vivo. Recently, bootstrap techniques have been introduced which overcome the necessity of a large number of scan repetitions. So far, the performances of these algorithms have only been analyzed using Monte Carlo simulations. The present work investigates and compares the predicted measurement precision of two model-based and two non-model-based bootstrap algorithms experimentally and demonstrates that using this approach diffusion sequences can be effectively optimized.