Use of capillaries in the construction of an MRI phantom for the assessment of diffusion tensor imaging: demonstration of performance

Research output: Contribution to journalArticlepeer-review

42 Scopus citations

Abstract

Although diffusion tensor imaging (DTI) shows great potential for the diagnosis of a variety of pathologies, no consensus for an appropriate assessment standard of DTI exists. This study examined the feasibility of using water-filled arrays of glass capillaries to construct a DTI phantom suitable for making repeated and reproducible measurements required in a quality assessment program. Three phantoms were constructed using arrays of capillaries with three inner diameters (23, 48, and 82 μm). Data were acquired using DTI protocols; the fractional anisotropy (FA), mean apparent diffusion coefficient (〈ADC〉) and principal eigenvectors of the diffusion tensors were calculated. This study demonstrated four results: (1) echo-planar images show that susceptibility within the capillary arrays does not lead to substantial differences in precessional frequency in regions containing the arrays and neither do the regions show noticeable image distortion; (2) principal eigenvectors of the diffusion tensors agree to within <10.3° of the array orientations; (3) mean FA values (0.18-0.50) and 〈ADC〉 values (1.40-1.93×10-3 mm2/s) within specified regions of interest are in general agreement with simulations after a simple noise correction; and (4) these array performance characteristics are observable using a typical clinical DTI protocol.

Original languageEnglish (US)
Pages (from-to)1349-1361
Number of pages13
JournalMagnetic Resonance Imaging
Volume24
Issue number10
DOIs
StatePublished - Dec 1 2006

Keywords

  • Anisotropy
  • Diffusion
  • MRI
  • Phantom
  • Tensor

ASJC Scopus subject areas

  • Biophysics
  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Use of capillaries in the construction of an MRI phantom for the assessment of diffusion tensor imaging: demonstration of performance'. Together they form a unique fingerprint.

Cite this