Nicholas P. Blockley1, Richard B. Buxton1
1Center for Functional MRI, University of California San Diego, La Jolla, CA, United States
Measurement of the BOLD response to hyperoxia has been proposed as a way to measure venous cerebral blood volume. We simulated two methods of achieving this aim with a detailed BOLD signal model, testing each for sensitivity to intersubject variations in haematocrit and baseline oxygen extraction. The first was the original method of normalising the tissue signal response to hyperoxia by the signal response of a large vein. The second was a new method based only on the tissue signal response. Compared with the original approach, this new method was found to be much less sensitive to intersubject variability.