Optimization of Diffusion-Tensor MRI for Tractography

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): The development of accurate, noninvasive methods for mapping white matter fiber-tracts in relation to brain pathologies is a goal of critical importance to the neurosurgical community. However, conventional fiber- tracking is based on spin-echo echo-planar diffusion tensor image acquisitions (SE-EPI-DTI), which suffer from BO-related image distortions and artifacts. Thus, a large percentage of white matter fiber-bundles are distorted, and/or terminated early, while others are completely undetected. This severely limits the clinical potential of fiber-tracking techniques. In contrast, PROPELLER-DTI (Periodically Rotated Overlapping ParallEL Lines with Enhanced Reconstruction) is relatively immune to BO-related artifacts. However, the imaging time in PROPELLER-DTI is markedly longer than in conventional SE-EPI-DTI. The broad objective of this research is to minimize noise and image artifacts in an accelerated version of PROPELLER-DTI (Turboprop-DTI), in order to provide a data acquisition and reconstruction technique that is optimized for tractography in a clinical setting. Our specific aims are to: (i) Incorporate physiological information in Turboprop-DTI image reconstruction in order to reduce noise, (ii) Develop improved reconstruction methods for Turboprop-DTI to further reduce noise and image artifacts, (iii) Develop optimized Turboprop-DTI acquisition schemes with clinically acceptable imaging time, and (iv) Compare tractography results obtained from SE-EPI-DTI and optimized Turboprop-DTI datasets. The accuracy and reproducibility of fiber-tracking results depend greatly on the noise levels and amount of artifacts of the DTI images. This study will address these issues and develop a PROPELLER-based imaging method that is optimized for white matter fiber- tracking in a clinical setting. The successful completion of this research will accelerate the development of fiber-tracking as a routinely used and powerful diagnostic tool.