RÉSUMÉ
The advent of parallel MRI over recent years has prompted a variety of techniques for performing parallel imaging. The main distinguishing feature among these is the specific way of solving the problem of image reconstruction from undersampled multiple-coil data. The clearest distinction in this respect is that between k-space method and image-domain method. The present paper reviews the basic reconstruction approaches, aiming to emphasize the common principles along with actual differences. To this end the treatment starts with an elaboration of the encoding mechanisms and sampling strategies that define the reconstruction task. Based on these considerations, the distinction between k-space and image-domain approaches is given. At the close of this paper are presented discussioins concerning noise propagation and control in parallel imaging and an outlook upon key issues to be addressed in the future.