Back

Transcranial in-depth ultrafast ultrasound brain imaging in non-human primates

Orset, T.; Royo, J.; Tanter, M.; Deffieux, T.; Thiebaut de Schotten, M.; Pouget, P.

2025-08-07 neuroscience
10.1101/2025.08.05.667956 bioRxiv
Show abstract

Ultrasound Localization Microscopy (ULM) offers sub-diffraction resolution vascular imaging by tracking individual contrast microbubbles. However, transcranial implementation in non-human primates remains challenging due to the skulls acoustic attenuation and distortion. This study presents a method for non-invasive transcranial ULM imaging of the anesthetized squirrel monkey through intact skull and skin. We systematically evaluated key ultrasound acquisition parameters and identified an optimal configuration balancing resolution, depth, and vascular representation. Using these refined parameters, we achieved a system resolution of 9.5 {micro}m, measured by Fourier Ring Correlation (FRC), and consistently visualized vessels as small as 23 {micro}m. Full-brain coronal and sagittal acquisitions revealed major vascular landmarks that were reproducible across animals and slices. Importantly, this resolution enables reliable detection of penetrating arterioles and venules, key elements in neurovascular coupling and implicated in various cerebrovascular pathologies. Finally, by tracking microbubbles, we quantified flow velocities from 8.8 to 55 mm/s, corresponding to flow rates ranging from 0.1 to 279 {micro}L/min, demonstrating the techniques ability to access micro- and macrovascular dynamics. This work establishes a framework for high-resolution, non-invasive brain imaging in non-human primates, and opens promising perspectives for studying cerebral structure, function, and dysfunction in translational neuroscience.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.