<b>Novel Processing Techniques for Unveiling Cerebral Physiological Pulsations in fMRI</b>

<p dir="ltr">Physiological signals in the brain, such as low-frequency oscillations, breathing, and heartbeat, are important for brain health. These signals affect blood flow, brain fluid movement, and waste clearance. Accurate cerebral vessel segmentation could facilitate fluid dynamics research in fMRI. However, most methods that find blood vessels use structural MRI. These methods often do not work well in fMRI because of misalignment caused by fMRI distortion. Also, most fMRI scans use a slow sampling rate (TR > 0.8 s). This slow rate causes the heartbeat signal to mix with lower frequencies. To address the challenge in vessel detection, large cerebral arteries and the superior sagittal sinus (SSS) are directly identified in fMRI space by robustly leveraging these vessels’ distinct pulsatile signal patterns during the cardiac cycle. This method was validated for accuracy in a local dataset with ground truth vessel segmentation and demonstrated high reproducibility with the Human Connectome Project-Aging dataset on 422 participants with repeated scans. After finding these regions, a novel region-specific hypersampling method was developed to resolve the physiological signals. This method uses fast cross-slice sampling within each TR to hypersample the fMRI signal, which selectively enhances the effective temporal resolution of fMRI signals within coherently pulsating neurofluid and tissue. This approach was applied to the publicly available HCP-A dataset (ages 36-90), increasing the resolvable frequency, which enables clear separation among the cardiac, respiration, and LFO oscillations. These two methods contribute to uncovering hidden physiological pulsations to advance the understanding of brain physiology and disease-related alterations.</p>