Bridge Develops a Noninvasive Ultrasound Brain-Computer Interface

Bridge is developing a noninvasive brain-computer interface (BCI) that uses ultrasound to record brain activity. The company has begun its first clinical study in volunteers in Redwood City, California.

Merge Labs and Bridge’s Brain-Computer Interface Research

Merge Labs was spun out of Forest Neurotech in January and counts Norman and Afflalo as co-founders, along with Altman, researcher Mikhail Shapiro, and technology entrepreneurs Alex Branier and Sandro Herbig.

Because ultrasound measures brain activity indirectly through blood flow rather than directly from neurons, there is a natural delay of several seconds between neuron firing and subsequent changes in blood flow and volume, said Mariam Shanesi, a neuroengineer at the University of Southern California.

“It captures slow fluctuations in neural activity,” Shanesi said. “This is fine for many applications, but it depends on what you want to build.”

How Ultrasound Brain-Computer Interfaces Could Work in Real Time

For real-time, fast-paced tasks, such as moving a robot’s limbs instantaneously or typing rapidly, this delay could challenge ultrasound-based brain-computer interfaces. Bridge plans to use predictive artificial intelligence models to reduce latency and improve interface performance.

Biedermann said the company is interested in pursuing both consumer and medical applications, but declined to comment on specific plans.

Bridge Begins Its First Clinical Study

Bridge’s first clinical study is observational, non-therapeutic, and non-diagnostic. It involves healthy adults and people who have undergone skull surgery. During the study, a small ultrasound probe is placed on the side of the participant’s head.

The probe uses a semiconductor silicon chip made by portable ultrasound company Butterfly Network to record brain activity while participants listen, see, move, and speak.

A Headphone-Like Device Designed for the Temporal Window

Bridge has not released images of the device, but Biedermann said it is “more of a headphone” that sits in the temporal window just above the ear. Because this is the thinnest part of the skull, “this is the easiest place to image,” Biedermann said.

The location is important because bone scatters and absorbs sound, creating technical challenges for noninvasive ultrasound imaging through the skull.

The Challenge of Imaging the Brain Through the Skull

In a 2024 survey, researchers at the California Institute of Technology and the University of Southern California demonstrated that an ultrasound-based brain-computer interface could read and decipher human brain activity. However, the approach required removing part of the skull and replacing it with a “window” through which ultrasound waves could pass.

Creating a system that can image brain activity noninvasively through the skull is one of the biggest technical hurdles Bridge faces.

“My vision focuses on non-invasive interfaces,” Biedermann said. “I think that’s where we have the greatest opportunity to help people.”