Meeting Banner
Abstract #2297

A Novel Approach to the Quantitative Estimation of the Oxygen Metabolism Response to a Neural Stimulus That Accounts for Uncertainty Due to Unmeasured Physiological Parameters

Aaron Benjamin Simon1, Valerie E M Griffeth1, Richard B. Buxton2

1Bioengineering, University of California San Diego, La Jolla, CA, United States; 2Radiology, University of California San Diego, La Jolla, CA, United States

We examined whether we could estimate the CMRO2 response to a stimulus, as well as the uncertainty of that estimate, by measuring baseline R2' and the CBF and BOLD responses. We related the three measurements to the CMRO2 response through a detailed, multi-compartment BOLD model and adopted a Bayesian approach to determine how uncertainty in unmeasured physiological parameters of the model affected the precision of the estimate the CMRO2 response. We tested the method with a hypercapnia challenge and found that the estimate agreed well with previous work but that the unmeasured parameters produced considerable uncertainty in the estimate.

Keywords

accomplished account accounts addition affected allow analyzed around arterial assess attributable audience bioengineering blood bold breathing calculation calibrated calibration capillary cerebral challenge chosen classic combinations combining compartmental compartments complete completely consisting constants containing corrected curve decay derived described detailed determine diffusion distribution distributions drawn dual essentially estimating estimation examined example experiment exponential extraction field finding finite finitely fluctuations formula fraction fractional generated good hypothesizing ignoring implementation importantly interval known macroscopic made many matrix measure measured measuring metabolism metric minute model models modified natl negligible neural novel oxygen period physiological posterior practical precisely precision principle prior produced product proof quantitative quantitatively quantity radiology random readout recent recovery reflect related relationships researchers response responses samples sampling series side significantly simple simplification simplified simultaneous slices slope stack stimulus studies subjects subtraction suggest surround taking target theoretical thick thresholding tissue typically uncertain uncertainty underlying unfixed uniform unmeasured useful variability variables venous volume volumes water whether