Detection and Characterization of Physiological Network Interactions in Pulsatile Motion of Cranial Blood Vessels Using Real-Time MRI

Thorge von der Ohe¹Vitali Telezki²Sabine Hofer²Peter Dechent²Martin Uecker³Mathias Bahr²Stefan Luther¹Ulrich Parlitz^1*

• ¹Max Planck Institute for Dynamics and Self-Organization, Göttingen, Germany
• ²Universitatsmedizin Gottingen Klinik fur Neurologie, Göttingen, Germany
• ³Institute of Biomedical Imaging, Technische Universitat Graz, Graz, Austria

We present a robust method to assess pulsatile motion of larger cranial blood vessels in the human brain from high spatio-temporal resolution real-time magnetic resonance (MR) imaging data. Together with percentile based thresholding in combination with a border detection algorithm and other empirical selection criteria, we are able to extract area-time series from pulsatile motion of blood vessels. In a proof of concept, we apply our method to the left and right vertebral arteries in a cohort of healthy subjects and extract heart and breathing rates from their pulsatile motion. While comparison to mean physiological reference values measured simultaneously with a photoplethysmogram and a breathing belt show no differences within the scope of the measurement accuracy, intra-subject differences for breathing rates detected in the left and right vertebral artery are high but not significant. Our findings suggest that our presented method is suited to access arterial pulsations of larger cranial vessels driven by heart or breathing rates, as part of the complex physiological network of heart and brain interactions.