Meeting Banner
Abstract #4229

A GA Guided K-Space Sampling for Compressed Sensing MRI

Hua Wang1, Yeyang Yu1, Bing Keong Li1, Adnan Trakic1, Mingjian Hong2, Feng Liu1, Stuart Crozier1

1The University of Queensland, Brisbane, QLD, Australia; 2ChongQing University, ChongQing, China

In this work, we presented a method to optimise the k-space sampling scheme for CS-MRI. The problem was simplified by treating the variable density functions in parts, and optimising the weighting factors with GA. A 2D brain and a 3D apple MRI image reconstruction illustrates an improved imaging quality with this method.

Keywords

ability accelerated according achieve achieved adjusted ameliorate apple applied applying assigned basis better binary blue brain chart chosen chromosome clarity close compressed compressible computationally conditions configure considering council crossover demanding density details determine distribute distributed distribution employing encoded evaluate even experimental exploit explored extent fact finished fitness fixed flow form framed freedom frequency function functions generated genetic grid gridded grids guided idea ideally illustrates implemented improve incoherence incoherent individually initial investigate iteration iterative linear localized look maintaining manner many matrix maximized mimic mutation next noise object optimal optimized parallel pattern patterns peak performance plot plots plotted population potentials practical problem procedure process proposed quality random randomly real reality reconstructed reconstruction redistribute reduce reduced reduction redundancy regard requirements resolutions respectively sampling scenarios scheme search selection sensing sharper simplified solution solutions space sparse sparsity still studies style support survive synthesized system termination tiled transform transparency treating validation variable variables variants various view weightings