Meeting Banner
Abstract #2395

An Improved Algorithm for the Estimation of Multi-Component T2 Distributions

Kelvin J. Layton1, 2, Mark Morelande1, Peter M. Farrell1, Bill Moran1, Leigh A. Johnston1, 3

1Electrical and Electronic Engineering, University of Melbourne, Melbourne, Victoria, Australia; 2National ICT Australia, Melbourne, Victoria, Australia; 3Florey Neuroscience Institutes (Parkville), Melbourne, Victoria, Australia

The underlying distribution of relaxation times contributing to an MR signal can provide useful information about the structure of brain tissue. Recently, this distribution has been modelled by discrete T2 components, which are estimated using a gradient-based least-squares optimisation algorithm. In this work, we demonstrate that this algorithm highly dependent on initialisation and SNR levels, often producing inaccurate estimates of the T2 components. We propose a Bayesian algorithm that overcomes these limitations and provides reliable T2 estimates at clinically achievable SNR. The algorithm is demonstrated through simulated and experimental data.

Abstract PDF Poster PDF

Keywords

allows alternatively although apart apply approaches approximate approximating approximation assumption away axial biological brain calculations challenge clinical clinically closer coil component components compute consider considered consistent consisting contain contributing converge conversely correcting correction cost decreased degradation described develop discrete diseases display displays distribution distributions echoes electrical electronic empirical engineering equation error errors estimating estimation every examine exhibits experimental exponentials fails fast find fitting flat flattened flattening function fusion generated global gradient include increasing info inspection institutes iteration iteratively kelvin known least likelihood likewise limitations local located maps mark marked matrix mode model moderate modes naive narrow national noise numerically overcomes parametric performance peter pixel pixels posterior prior problem procedure progressive proposed random rather realistic recently recovering reduced relatively reliability reliable repeated represent represented reveal robust sclerosis sharp sheep simulated simulation slice slow solution square squared squares starting structure subsequently theory though tissue tractability transmit true typical update utility variable white wide