Abstract #0435

Jones D

The bootstrap is a powerful tool in DT-MRI tractography, allowing assignment of confidence to reconstructed pathways without any prior assumptions about the data. It is unlikely to find widespread acceptance, however, due to the long acquisition times that are needed for robust bootstrapping. We discuss the application of an alternative to conventional bootstrapping Wild Bootstrapping, to fiber tracking. This allows extraction of probability distributions from data collected in a fraction of the time that conventional bootstrapping requires. We show good agreement between the two methods, indicating that previously published bootstrap-dependent methods may be brought within the clinical domain.