IMAGE QUALITY OF GRAPPA ACCELERATED EPI IMAGE QUALITY

点击翻译按钮获取中文摘要

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Introduction. It has been demonstrated that GRAPPA1 and SENSE2 are extremely powerful methods to reduce geometric distortions and image blurring in echo planar imaging (EPI) 3, 4. This bears great potential to improve image quality and overall lesion conspicuity for numerous clinical applications. Further image degradation in EPI is due to slight imbalances between odd and even echoes that lead to echo misalignment and hence FOV/2 ghosts. The objective of this study is to show that for a self-calibrated parallel imaging reconstruction approach even slight inaccuracies in EPI phase correction can lead to substantial reconstruction errors. Materials and Methods. Several factors, such as gradient imbalances, filter delays, and eddy currents, are causative for errors between odd and even echoes in EPI. These errors typically cause phase errors and echo misalignment in k-space and therefore result in FOV/2 ghosts in single-shot EPI scans; ghosts of higher spatial periodicity occur for interleaved EPI. Based on a reference scan, zero and first order phase correction is usually applied to the data in the image domain prior to gridding to avoid these ghosts. Here, we demonstrate that inaccuracies in EPI phase correction can lead to unexpectedly high image degradation even if the ghosts appear to minor in the original b=0 data. Although any ghost correction scheme that minimizes the ghosting artifacts should be applicable, we have used a referenceless method that was recently developed in our lab. This phase correction is based on image based entropy to determine the appropriate first and zero order phase terms to minimize the amount of FOV/2 ghosting in a rapid and reliable fashion. References. 1Griswold, M. et al. MRM 47: 1202-10, 2002; 2Pruessmann, K. et al. MRM 42: 952-62, 1999; 3Bammer, R. et al. MRM 46: 548-54, 2001; 4Heidemann, R. et al. 9th Annual Meeting of the ISMRM, Glasgow, Scotland, 2001, p 169. 5Skare, S et al. ISMRM DWI workshop, Lake Louise: p 16:2005 Acknowledgements. This work was supported in part by the NIH (1R01EB002771), the Center of Advanced MR Technology at Stanford (P41RR09784), Lucas Foundation, and Oak Foundation.