Akademik Veri Yönetim Sistemi
European Biotechnology Conference, Letonya, 5 - 07 Mayıs 2016, cilt.231
K-space in magnetic resonance imaging (MRI) is a raw data space that is used to store MR signals during data acquisition. Fourier Transform of the k-space yields the image. The sampling of k-space is commonly performed using Cartesian trajectories on most clinical scanners. In a standard two-dimensional (2D) Cartesian trajectory sampling, the k-space is divided into planes as many as imaging slices and each k-space plane consists of rows as many as radiofrequency (RF) pulses. Recently non-Cartesian sampling trajectories attract attention since they offer faster data acquisition. 2D/3D stacks of spirals and 3D shells could be given as examples for non-Cartesian trajectories.
We developed 3D concentric shells sampling trajectories in MATLAB to implement on an MR scanner [1]. We tested that concentric shell sampling of k-space allows reduced number of RF pulses and efficient use of gradients, thus accelerates image acquisition. Without any under-sampling, shell-trajectories could offer about several-fold increased speed compared to 3D Cartesian acquisitions.