Image Reconstruction for a Rotating Radiofrequency Coil (RRFC) Using Self-Calibrated Sensitivity From Radial Sampling

• Published in: IEEE transactions on biomedical engineering 2017-02, Vol.64 (2), p.274-283
• Main Authors: Jin, Jin, Weber, Ewald, Tesiram, Yasvir, Hugger, Thimo, Li, Mingyan, Fuentes, Miguel, Ullmann, Peter, Stark, Simon, Junge, Sven, Liu, Feng, Crozier, Stuart
• Format: Article
• Language: English
• Subjects: Algorithms; Angular position; Angular velocity; Calibration; Coils; Compressed sensing (CS); Encoding; Equipment Design; GRAPPA; Image processing; Image Processing, Computer-Assisted - methods; Image reconstruction; Imaging; Imaging techniques; Interpolation; Magnetic resonance imaging; Magnetic Resonance Imaging - instrumentation; Magnetic Resonance Imaging - methods; Mathematical models; NMR; Nuclear magnetic resonance; Optical fibers; Optical measuring instruments; Optical transmitters; Phantoms, Imaging; Prototypes; Radio frequency; Recovery; regularized image reconstruction; rotating radiofrequency coil (RRFC); Rotation; SENSE; Sensitivity; sensitivity self-calibration; Signal Processing, Computer-Assisted; sparsity; Spokes; Velocity
• ISSN: 0018-9294; 1558-2531
• DOI: 10.1109/TBME.2016.2552489

The purpose of this study was to develop a practical magnetic resonance imaging (MRI) scheme for the latest rotating radiofrequency coil (RRFC) design at 9.4 T. The new prototype RRFC was integrated with an optical sensor to facilitate recording of its angular positions relative to the sequence timing. In imaging, the RRFC was used together with radial k-space trajectories. To recover the image, the radial spokes were grouped according to the coil locations. Using an Eigen-decomposition approach, an array of location-dependent sensitivity maps was extracted from the central regions of the segmented k-space, enabling parallel-imaging techniques for image recovery in a straightforward manner. When the RRFC angular velocity is carefully designed and accurately controlled according to the sequence timing, the encoding by means of varying RRFC sensitivity maps can be accurately calibrated for a faithful image recovery. Approximations were made to counteract the variations of the RRFC angular velocity, providing successful image reconstruction at 9.4 T. The current study demonstrated a new and practical imaging scheme for RRFC-MRI. It is able to extract the temporally varying sensitivity maps retrospectively from the k-space acquisition itself, without resorting to electromagnetic simulation or numerical interpolation. The proposed imaging scheme and the supporting engineering solutions of the RRFC prototype enable accurate image reconstructions. These new developments pave the way for routine applications of the RRFC, and bode well for its further development in providing simultaneous multinuclear imaging by incorporating, for example, independent X-nuclear coil elements into the rotating structure.