Translator Disclaimer
18 March 2015 Needle position estimation from sub-sampled k-space data for MRI-guided interventions
Author Affiliations +
MRI-guided interventions have gained much interest. They profit from intervention synchronous data acquisition and image visualization. Due to long data acquisition durations, ergonomic limitations may occur. For a trueFISP MRI-data acquisition sequence, a time sparing sub-sampling strategy has been developed that is adapted to amagnetic needle detection. A symmetrical and contrast rich susceptibility needle artifact, i.e. an approximately rectangular gray scale profile is assumed. The 1-D-Fourier transformed of a rectangular function is a sinc-function. Its periodicity is exploited by sampling only along a few orthogonal trajectories in k-space. Because a needle moves during intervention, its tip region resembles a rectangle in a time-difference image that is reconstructed from such sub-sampled k-spaces acquired at different time stamps. In different phantom experiments, a needle was pushed forward along a reference trajectory, which was determined from a needle holders geometric parameters. In addition, the trajectory of the needle tip was estimated by the method described above. Only ca. 4 to 5% of the entire k-space data was used for needle tip estimation. The misalignment of needle orientation and needle tip position, i.e. the differences between reference and estimated values, is small and even in its worst case less than 2 mm. The results show that the method is applicable under nearly real conditions. Next steps are addressed to the validation of the method for clinical data.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Sebastian Schmitt, Morwan Choli, and Heinrich M. Overhoff "Needle position estimation from sub-sampled k-space data for MRI-guided interventions", Proc. SPIE 9415, Medical Imaging 2015: Image-Guided Procedures, Robotic Interventions, and Modeling, 94152G (18 March 2015);


Back to Top