Meeting Banner
Abstract #2851

Regressor Selection for RETROICOR and RETROKCOR Corrections of 3D FMRI Data

Rob H.N. Tijssen1, 2, Mark Jenkinson1, Karla L. Miller1

1FMRIB Centre, Oxford University, Oxford, United Kingdom; 2University Medical Center Utrecht, Utrecht, Netherlands

In this work we explore the use of retrospective corrections, in order to remove temporal instabilities from 3D FMRI acquisitions. Using the Bayesian Information Criterion (BIC) to determine an optimal subset, previously proposed for RETROICOR, we aim to optimize the regressor set for various 3D acquisition methods (e.g., SPGR and bSSFP). The results suggest that the optimal number of regressors is highly dependent on the acquisition and can even be zero (i.e., no retrospective correction) in some cases (e.g., when a real-time cardiac synchronized readout is used to correct for cardiac fluctuations prospectively).

Keywords

acquisition acquisitions additionally advisable aims allow amount another appear applied appropriately arrow balanced beforehand bellows brooks calculating candidate cardiac channel coil combined computationally consider constructed corrected correction corrections correlated criterion defined degrees demanding dependent depending derivatives determine determined either employed expanding expansion expect explain explore favored fitting fluctuations fraction freedom functional glover greatest head healthy highly included indicating individual individually instabilities instead interactions iteration iteratively kingdom like list listed little location long mark marks matrix maximize meaningful medical miller mitigate mitigated model models next normalized nuisance optimal optimizing oxford penalized physiological pneumatic portion potential potentially prevent previously procedure prone proposed prospective prospectively randomly readout readouts real recorded reduce reduction regression regressors remaining remove residual resolution respiratory retrospective segmented selected selection series slice sorted space spatial spoiled squares statistics step striking subset successfully suggest synchronization synchronized table temporal terms theory third together trio variance varies various volumes volunteers waveforms whereas whether zero