Skip to content
Tech News
← Back to articles

Ultrasound imaging of the brain

read original more articles
Why This Matters

The development of ultrasound-based brain imaging represents a significant breakthrough in non-invasive neural interfaces, offering MRI-level detail without the need for invasive procedures. This technology could revolutionize brain research, medical diagnostics, and future brain-computer interfaces, making them more accessible and less risky for consumers and clinicians alike.

Key Takeaways

June 24, 2026

A few years ago, a paper came out that blew our minds. The idea was that you can decode what someone is looking at just from their brain activity.

Reconstructing seen images from brain activity — seen image (left) and reconstruction (right) for each (MindEye, decoded from fMRI)

It’s wild and shows just a glimmer of what a telepathic future would be like. Unfortunately, it requires an MRI machine, which sadly can’t be worn on the head.

In fact, the first bottleneck to the whole field of mind interfacing is the hardware. There are currently two extremes: drill a hole through your skull and stick electrodes in your brain, or record blurry-at-best images of brain activity outside the head with EEG.

We’ve been building a new type of hardware that requires no drilling, and gives you MRI-level detail of the brain.

It’s based on ultrasound. It exploits a connection between your vascular system and your neurons — when neurons fire, more blood is delivered to the neurons. We send ultrasound waves through the skull, and they scatter off red blood cells. We can then form maps of blood flow and volume throughout the brain.

Ultrasound propagating through the human head.

We think there are two requirements in a general-purpose mind interface. The first is that it has to be able to see a large part of the brain. Even with 1000 electrodes, you capture at most 0.001% of the brain. This is great for a narrow task like controlling a cursor. But thoughts are distributed all over the brain.

The second requirement is detail, or resolution. Modalities like EEG and MEG have great field of view, but capture blurry images of brain activity. This is fundamental, it’s due to the way electric and magnetic fields propagate, and this is not solved by scaling to millions of sensors.

... continue reading