kspaceencoding

kspace encoding refers to the process of acquiring magnetic resonance imaging (MRI) data. In MRI, images are not directly sampled as a picture but rather as a representation in a frequency domain called k-space. k-space contains information about the spatial frequencies of the MR signal. The process of filling k-space, or k-space encoding, dictates how the spatial information is collected and, consequently, the final image quality.

Typically, k-space is filled line by line. Gradient magnetic fields are used to manipulate the phase of the MR signal, allowing for the spatial encoding of different parts of the object being imaged. One common method is the Cartesian acquisition, where k-space is filled in a raster-scan pattern. Other techniques, such as radial or spiral trajectories, sample k-space in non-Cartesian patterns, offering advantages in specific applications like fast imaging or imaging of moving structures. The data collected in k-space is then Fourier transformed to reconstruct the anatomical image. The rate at which k-space is sampled influences the spatial resolution and the time it takes to acquire the image. Undersampling k-space, for instance, can accelerate acquisition but may lead to artifacts if not handled appropriately with advanced reconstruction algorithms.