Research On MRI PROPELLER Algorithm And Motion Artifact Elimination

Magnetic Resonance Imaging(MRI)is a new medical imaging method, it makes images based on nuclear magnetic resonance signals in the static magnetic field. It has many features such as high image resolution, multi-imaging Parameters, non-invasive and arbitrary slice imaging, No ionizing radiation, et al. Now MRI is one of the most advanced medial imaging methods, and is more and more widely used in the clinical applications. However, MRI is sensitive to motion; in imaging processing the patient’s mild movement can fuzz the image and seriously influence the medical diagnosis. Therefore, How to effectively overcome the effect of motion artifact, has been the hot spot and one of the technical problem in MRI. In this paper, we carried out the research on motion artifact elimination based on the Periodically Rotated Overlapping Parallel Lines Enhanced Reconstruction (PROPELLER).After analyzing the formation mechanism and the correction methods of the motion artifact, we do in-depth study on PROPELLER algorithm. When perform rotation estimation using PROPELLER algorithm, the amplitude information are used in translation but not the phase information which account for the drop of accuracy and robustness. Therefore, an improved algorithm for motion parameter estimation is proposed which using RMI as the measure. Motion parameters are searched by the improved algorithm through the temporary image registration of each strip, and are estimated in two steps in the image domain and the frequency domain. The simulation results show that, for different imaging organization, the improved algorithm can correct motion artifact effectively, obtain more clarity reconstructed image.The conventional PROPELLER estimates and corrects the motion parameters using the resample data in center area of k-space only, which search and correction the motion parameters using the low frequency data in the low resolution. Not only wastes the data outside the center area, but also reduces the motion parameters searching precision. Therefore, the improved PROPELLER based on wavelet transformation will be proposed in this paper. The improved algorithm reconstructs a set of temporary images of all k-space strips firstly, and then searches the motion parameters of each k-space strip by using the normalized mutual information as the similarity measure in the multi-resolution. Finally, corrects motion artifacts. The algorithm can reduce the estimation error which is caused by the little quantity of the data. Through the analysis of experimental results, the improved method can effectively search the motion parameters, remove artifact, and improve the precision of the image reconstruction.