首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Anisotropic Phantoms for Quantitative Diffusion Tensor Imaging and Fiber-Tracking Validation
Authors:R Lorenz  M E Bellemann  J Hennig  K A Il'yasov
Institution:(1) Department of Diagnostic Radiology, Medical Physics, University Hospital Freiburg, Germany;(2) Department of Biomedical Engineering, University of Applied Sciences, Jena, Germany;(3) Physics Department, Kazan State University, Kazan, Russia
Abstract:Different fiber materials (hemp, linen, viscose rayon, polyamide and dyneema twine) were tested for their suitability as fiber phantoms for diffusion tensor imaging (DTI) calibration on clinical magnetic resonance imaging systems with common diffusion-weighted echo planar imaging sequences. Additionally, the potential for fiber tracking validation of these fiber phantoms was investigated. For phantom manufacturing the fibers were wound up into a bundle of parallel fibers enwrapped by plastic ribbon. The most homogenously distributed fractional anisotropy (FA) values (0.63 ± 0.10) were determined in the dyneema and polyamide (0.3 ± 0.1) fiber phantom. FA values in the viscose, linen and hemp bundles were at high variations (about 0.2 ± 0.10). The dispersion of the direction of the principal eigenvector in the polyamide and dyneema phantom was less than 7°, for the other fiber phantoms it was over 30°. Thus, the presented results may indicate that polyamide- and dyneema-based fiber phantoms provide the opportunity for verification and validation of DTI sequences on clinical scanner. Additionally, they can be applicable for testing the accuracy of fiber tracking algorithms. A strong parallel alignment of the fibers with a constant compression grade of the fiber bundles could be achieved by machine-made production. This could also provide highly reproducible diffusion properties within the anisotropic fiber phantoms. Authors' address: Kamil A. Il'yasov, Physics Department, Kazan State University, Kremlevskaya ulitsa 18, Kazan 420008, Russian Federation
Keywords:
本文献已被 SpringerLink 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号