Quantification of oxygen metabolic rates in Human brain with dynamic 17 O MRI: Profile likelihood analysis

PURPOSE Parameter identifiability and confidence intervals were determined using a profile likelihood (PL) analysis method in a quantification model of the cerebral metabolic rate of oxygen consumption (CMRO2 ) with direct 17 O MRI. METHODS Three-dimensional dynamic 17 O MRI datasets of the human brain were acquired after inhalation of 17 O2 gas with the help of a rebreathing system, and CMRO2 was quantified with a pharmacokinetic model. To analyze the influence of the different model parameters on the identifiability of CMRO2 , PLs were calculated for different settings of the model parameters. In particular, the 17 O enrichment fraction of the inhaled 17 O2 gas, α, was investigated assuming a constant and a linearly varying model. Identifiability was analyzed for white and gray matter, and the dependency on different priors was studied. RESULTS Prior knowledge about only one α-related parameter was sufficient to resolve the CMRO2 nonidentifiability, and CMRO2 rates (0.72-0.99 µmol/gtissue /min in white matter, 1.02-1.78 µmol/gtissue /min in gray matter) are in a good agreement with the results of 15 O positron emission tomography studies. Nonconstant α values significantly improved model fitting. CONCLUSION The profile likelihood analysis shows that CMRO2 can be measured reliably in 17 O gas MRI experiment if the 17 O enrichment fraction is used as prior information for the model calculations. Magn Reson Med 78:1157-1167, 2017. © 2016 International Society for Magnetic Resonance in Medicine.