Estimating blood oxygen saturation through susceptibility source separation: A new standpoint on quantitative BOLD...

Area of Science:

• Biomedical Imaging
• Neuroscience
• Medical Physics

Background:

• Current MRI methods for estimating cerebral blood oxygen saturation are limited by neglecting extravascular magnetic susceptibility sources.
• Existing quantitative BOLD and quantitative susceptibility mapping (QSM) approaches fail to account for the geometric arrangement of these extravascular sources, impacting signal decay accuracy.

Purpose of the Study:

• To develop and validate a new MRI model that differentiates intravascular and extravascular magnetic susceptibility contributions.
• To generate accurate blood-oxygen saturation maps and non-vascular susceptibility maps by considering both signal magnitude and phase.

Main Methods:

• A novel MRI model was developed to analyze signal contributions from intravascular and extravascular compartments in gradient echo sequences.
• The new model and previous MRI techniques were applied to rat data from healthy and stroke models.
• Comparisons were made across anatomical regions with varying extravascular magnetic susceptibilities.

Main Results:

• The proposed model demonstrated improved blood oxygen saturation estimations, particularly in myelin-rich white matter regions.
• The model successfully delineated decreased oxygen saturation in stroke lesions following middle cerebral artery occlusion.
• Incorporating extravascular susceptibility source characterization significantly enhanced overall results.

Conclusions:

• The novel MRI model provides more accurate estimations of cerebral blood oxygen saturation by including extravascular sources.
• Accurate characterization of extravascular magnetic susceptibility is crucial for advancing MRI in pathological applications, such as stroke imaging.