A team called Aleph has captured the most detailed vascular map of a living human brain using transcranial ultrasound. By sending high‑frequency pulses through an intact skull and tracking scattered echoes from red blood cells, they produced the first 3D image of ultrasound localization microscopy in a human. The scan resolves large vessels, pial arteries and arterioles, a level of detail the authors say rivals MRI.

The resolution stems from a super‑resolution trick: sparsely injected FDA‑approved microbubbles act as point scatterers. Each bubble appears as a wavelength‑wide blur, but sub‑pixel fitting pins its centre far below the diffraction limit. Accumulating millions of positions yields a volumetric image 100 times finer than comparable CT. The process runs over a four‑minute infusion, generating terabytes of raw data.

Aleph released the entire processing pipeline and dataset, inviting researchers to explore vascular signatures of stroke, Alzheimer’s and traumatic brain injury. Falling hardware costs—thanks to firms like Butterfly, which now sells smartphone‑sized probes for under $100 k—make widespread use plausible. The authors argue that end‑to‑end machine learning on these massive streams could eventually eliminate contrast agents, delivering high‑resolution, non‑invasive neurovascular imaging today.