Improved spatial localization based on flow-moment-nulled and intra-voxel incoherent motion-weighted fMRI.

Published

Journal Article

Functional MRI signal based on the blood oxygenation level-dependent contrast can reveal brain vascular activities secondary to neuronal activation. It could, however, arise from vascular compartments of all sizes, and in particular, be largely influenced by contributions of large vein origins that are distant from the neuronal activities. Alternative contrasts can be generated based on the cerebral blood flow or volume changes that would provide complementary information to help achieve more accurate localization to the small vessel origins. Recent reports also indicated that apparent diffusion coefficient-based contrast using intravoxel incoherent motion (IVIM) weighting could be used to efficiently detect synchronized signal changes with the functional activities. It was found that this contrast has significant arterial contribution where flow changes are more dominant. In this study, a refined approach was proposed that incorporated the flow-moment-nulling (FMN) strategy to study signal changes from the brain activation. The results were then compared with those from conventional IVIM- and BOLD-weighted acquisitions. It was shown that the activated region using the new acquisition strategy had smaller spatial extent, which was contained within the activated areas from the other two methods. Based on the known characteristics of the conventional IVIM and BOLD contrasts, it was inferred that the FMN-IVIM acquisition had improved selective sensitivity towards smaller vessels where volume changes were prevalent. Therefore, such an acquisition method may provide more specific spatial localization closely coupled to the true neuronal activities.

Full Text

Duke Authors

Cited Authors

  • Song, AW; Li, T

Published Date

  • May 2003

Published In

Volume / Issue

  • 16 / 3

Start / End Page

  • 137 - 143

PubMed ID

  • 12884357

Pubmed Central ID

  • 12884357

International Standard Serial Number (ISSN)

  • 0952-3480

Digital Object Identifier (DOI)

  • 10.1002/nbm.819

Language

  • eng

Conference Location

  • England