Meeting Banner
Abstract #1561

QSM: fast selection of optimal regularization weights

Job Gijsbertus Bouwman1 and Peter R Seevinck1

1Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands

Quantitative Susceptibility Mapping reconstructions may benefit from L1-regularization and magnitude weighing, however these iterative reconstruction methods are time-consuming. Recently, progression has been made in reducing the reconstruction times with Split Bregman iterations, allowing subject-specific regularization weights. Here a further reduction of the reconstruction time is reported, mostly based on accelerating the automatic selection of the optimal regularization parameter. The overall procedure reduces computational load more than threefold, without accuracy loss. Reduction of reconstruction times, may contribute to realize QSM algorithms which are either clinically feasible, or that may pave the way to include more sophisticated regularization mechanisms.

This abstract and the presentation materials are available to members only; a login is required.

Join Here