Caltech scientists say they can read the human brain with ultrasound

A new technique uses exactly the same ultrasound imaging – the same type that allows parents to see their baby before birth – to read and even predict activity in the brain. In an interesting study published in the journal Monday, Caltech scientists were able to use ultrasound to hear because of the blood circulating in different parts of the brain, which they quickly realized was a proxy for which neural regions were active at any moment, an interesting study published in the journal Monday Called neurons.

After running data from a primate study into the algorithm, they further learned that specific patterns of blood flow not only matched but also predicted what action the primate would take and when they were going to do it – and there was no reason why people should not work. Overall, predictive neural imaging could usher in a new era of brain-computer interface technology that is more accurate and less dangerous than other tools on the market, scientists claimed in a press release.

“The first milestone was shown that ultrasound can capture brain signals related to the thought process of physical movement,” lead study author David Maresca said in the release. The use of ultrasound can solve a major problem in the world of neural imaging and brain-computer interfaces. On the one hand, we have implants and electrodes that can take very precise recordings of the brain, but they only used in extreme cases in patients with severe epilepsy because they require surgical treatment of invasive and potentially harmful brain.

On the other hand, we have noninvasive brain imaging tools like functional MRI or EEG arrays, but they either give gap readings or require huge, almost house-sized machines. Ultrasound, though, seems to offer the best of both. Scientists can use it to visualize the brain in the size of 100 nanometers – just ten individual neurons or the size of a human hair – and it does not require brain surgery. “Functional ultrasound imaging manages to record these signals with 10 times more sensitivity and better resolution than effective MRI,” Maresca said in a statement. “This finding is the key to the success of brain-machine interfacing based on functional ultrasound.”