Functional magnetic resonance imaging (fMRI) is driving a revolution in the brain sciences, providing new insights into the brain's normal functional organization and its alteration during disease. However, it is becoming clear that hemodynamic responses measured by fMRI can have a large variation between and within subjects as blood flow and oxygenation are influenced by factors other than the underlying neuronal and metabolic processes. Thus, the utility of fMRI would be improved if it provided more direct measures of brain activation. One important effort in this direction is using the hemodynamic measures to estimate the cerebral metabolic rate of oxygen (CMRO2), a metabolic marker that is more directly coupled to brain activation in health and disease. However, the estimation is dependent on a model of the vascular response to neuronal and metabolic signals. The hemodynamic response to brain activation is driven primarily by active arteriolar dilation and oxygen consumption, for which a qualitative biophysical model is conceptually straightforward and easily used to estimate CMRO2. Utilization of such qualitative models of the vascular and oxygen transport responses are becoming more common in analyzing fMRI and optical data;however there has been little confirmation of the accuracy of the methodology. Guided by direct measures of arteriole dilation and oxygen consumption by advanced microscopic imaging methods and a detailed microscopic vascular anatomical network (VAN) model, we will develop a qualitative model based on the windkessel model to accurately estimate CMRO2. Our goal is to establish the accuracy of the methodology to set the foundation for its routine use with fMRI and optical imaging in basic science and clinical care. Advanced optical microscopy methods are central to new discoveries in cerebrophysiology through their ability to measure multiple physiological processes with sub-cellular resolution. A VAN model, based on these measurable quantities, is required to integrate results from multiple descriptive experiments and to enable a more rigorous and testable examination of the cerebrovascular physiology that scales from the microscopic to the macroscopic. We will advance our VAN model in concert with novel optical microscopy measurements of the cerebrovascular physiology to learn about parameters relevant to the estimate of CMRO2 in humans such as the compliance of different vascular segments, the oxygen permeability of arteriole, capillary, and venules walls, the oxygen efflux from the tissue, and tissue oxygen reserve. We will then utilize this VAN model to determine the accuracy of the lumped parameter windkessel model, which is more conducive to routine analysis of human brain imaging data.
Showing the most recent 10 out of 57 publications
