Robust Estimation of Venous Cerebral Blood Volume Using Flow-Diffusion Modeling Combined with R2' and Perfusion Measurements

Motivation: Factoring out venous cerebral blood volume (vCBV) and oxygen extraction fraction (OEF) from the relaxation of R2’ poses a significant challenge. Goal(s): We combined robust measurements of R2’ relaxation from GE-SE EPIK and perfusion from VSI-ASL with a flow-diffusion model of oxygen transport to improve estimation of vCBV and OEF. Approach: The proposed method was tested in normoxia, hyperoxia and hypercapnia against validated TRUST measurements of OEF. Results: Maps of OEF and vCBV accurately depicted the oxygen metabolism at baseline as well as during the gas challenges without the need of any calibration, with high resolution and within reasonable acquisition time. Impact: The integration of the oxygen flow-diffusion model with GE-SE EPIK measurements of R2’ relaxation and VSI-ASL measurements of perfusion gives high resolution parametric maps of oxygen metabolism within reasonable acquisition time, ultimately reducing the gap between research and clinical practise.