Integrating a flow-diffusion model of oxygen transport with qBOLD for contrast-agent- and gas-free mapping of deoxygenated CBV, OEF and CMRO2

Measuring brain oxygenation is crucial for understanding cerebral metabolism, as oxygen availability directly influences neuronal function. Despite its potential for investigating neurological conditions, oxygen consumption mapping with MRI is not routinely applied in clinical contexts due to the challenges in making rapid reliable measurements as methods often rely on contrast-agent injection or gas-challenges. Here we propose a new quantitative BOLD strategy to simultaneously estimate oxygen consumption and blood volume without contrast-agent or gas-challenge through the integration of a flow-diffusion model of oxygen transport: flow-diffusion qBOLD. This new method was compared to qBOLD estimates from contrast-agent-based blood volume in non-human primates and yielded high correlations across grey matter regions. Flow-diffusion qBOLD was then applied to human data and showed a good correlation trend with global TRUST oxygen extraction fraction measures. Comparisons with measurements from calibrated BOLD with breath-holding challenge highlighted variable agreement across grey matter region and enabled to identify the potential pitfalls of the newly proposed imaging framework. Last, cross-species comparisons of cortical grey matter variations in flow-diffusion qBOLD parameters provided preliminary resources to validate blood volume estimates without contrast-agent injection in humans. Overall, this strategy holds promise towards the clinical application of oxygen consumption estimation in challenging patient populations.