Method for improving the accuracy of quantitative cerebral perfusion imaging

Ken E. Sakaie, Wanyong Shin, Kenneth R. Curtin, Richard M. McCarthy, Ty A. Cashen, Timothy J. Carroll*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

55 Scopus citations


Purpose: To improve the accuracy of dynamic susceptibility contrast (DSC) measurements of cerebral blood flow (CBF) and volume (CBV). Materials and Methods: In eight volunteers, steady-state CBV (CBVSS) was measured using TrueFISP readout of inversion recovery (IR) before and after injection of a bolus of contrast. A standard DSC (STD) perfusion measurement was performed by echo-planar imaging (EPI) during passage of the bolus and subsequently used to calculate the CBF (CBFDSC) and CBV (CBVDSC). The ratio of CBVSS to CBVDSC was used to calibrate measurements of CBV and CBF on a subject-by-subject basis. Results: Agreement of values of CBV (1.77 ± 0.27 mL/100 g in white matter (WM), 3.65 ± 1.04 mL/100 g in gray matter (GM)), and CBF (23.6 ± 2.4 mL/(100 g min) in WM, 57.3 ± 18.2 mL/(100 g min) in GM) with published gold-standard values shows improvement after calibration. An F-test comparison of the coefficients of variation of the CBV and CBF showed a significant reduction, with calibration, of the variability of CBV in WM (P < 0.001) and GM (P < 0.03), and of CBF in WM (P < 0.0001). Conclusion: The addition of a CBVSS measurement to an STD measurement of cerebral perfusion improves the accuracy of CBV and CBF measurements. The method may prove useful for assessing patients suffering from acute stroke.

Original languageEnglish (US)
Pages (from-to)512-519
Number of pages8
JournalJournal of Magnetic Resonance Imaging
Issue number5
StatePublished - May 1 2005


  • Cerebral blood flow
  • Contrast enhanced
  • Magnetic resonance imaging
  • Perfusion
  • Stroke

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Method for improving the accuracy of quantitative cerebral perfusion imaging'. Together they form a unique fingerprint.

Cite this