Implant could restore lost senses, provide sensory feedback for prosthetic limbs

Roughly the size of a postage stamp and thinner than a credit card, the new device is less invasive than what had been developed previously by the team. Instead of extending into the brain through a tiny cranial defect, the new soft, flexible device conforms to the surface of the skull and shines light through the bone. Photo by Mingzheng Wu of the Rogers Research Group

In a new leap for neurobiology and bioelectronics, Northwestern University scientists have developed a wireless device that uses light to send information directly to the brain — bypassing the body’s natural sensory pathways.

The soft, flexible device sits under the scalp but on top of the skull, where it delivers precise patterns of light through the bone to activate neurons across the cortex.

In experiments, scientists used the device’s tiny, patterned bursts of light to activate specific populations of neurons deep inside the brains of mouse models. (These neurons are genetically modified to respond to light.) The mice quickly learned to interpret these patterns as meaningful signals, which they could recognize and use. Even without touch, sight or sound involved, the animals received information to make decisions and successfully completed behavioral tasks.

The technology has immense potential for various therapeutic applications, including providing sensory feedback for prosthetic limbs, delivering artificial stimuli for future vision or hearing prostheses, modulating pain perception without opioids or systemic drugs, enhancing rehabilitation after stroke or injury, controlling robotic limbs with the brain and more.

The study will be published in the journal Nature Neuroscience.

“Our brains are constantly turning electrical activity into experiences, and this technology gives us a way to tap into that process directly,” said Northwestern neurobiologist Yevgenia Kozorovitskiy, who led the experimental work. “This platform lets us create entirely new signals and see how the brain learns to use them. It brings us just a little bit closer to restoring lost senses after injuries or disease while offering a window into the basic principles that allow us to perceive the world.

“Developing this device required rethinking how to deliver patterned stimulation to the brain in a format that is both minimally invasive and fully implantable,” said Northwestern bioelectronics pioneer John A. Rogers, who led the technology development. “By integrating a soft, conformable array of micro-LEDs — each as small as a single strand of human hair — with a wirelessly powered control module, we created a system that can be programmed in real time while remaining completely beneath the skin, without any measurable effect on natural behaviors of the animals. It represents a significant step forward in building devices that can interface with the brain without the need for burdensome wires or bulky external hardware. It’s valuable both in the immediate term for basic neuroscience research and in the longer term for addressing health challenges in humans.”

Kozorovitskiy is the Irving M. Klotz Professor of Neurobiology in Northwestern’s Weinberg College of Arts and Sciences. She also is a member of the Chemistry of Life Processes Institute. Rogers is the Louis Simpson and Kimberly Querrey Professor of Materials Science and Engineering, Biomedical Engineering and Neurological Surgery in the McCormick School of Engineering and Northwestern University Feinberg School of Medicine. He also is the director of the Querrey Simpson Institute for Bioelectronics. Mingzheng Wu, a postdoctoral fellow in the Rogers and Kozorovitskiy laboratories, is the study’s first author.

The soft, flexible device sits under the scalp but on top of the skull, where it delivers precise patterns of light through the bone to activate neurons across the cortex.

Replicating natural patterns of brain activity

The new study builds off previous work, in which Kozorovitskiy and Rogers introduced the first fully implantable, programmable, wireless, battery-free device capable of controlling neurons with light. Published in Nature Neuroscience in 2021, the previous study used a single micro-LED probe to influence social behavior in mice. While previous optogenetics (a method for controlling neurons with light) research required fiberoptic wires, which restrained mice, the wireless version allowed the animals to look and behave normally in social settings.

The new iteration takes this research a step further by enabling richer, more flexible communication with the brain. Going beyond the ability to activate and deactivate a single region of neurons, the new device features a programmable array of up to 64 micro-LEDs. With real-time control over each LED, researchers can send complex sequences to the brain that may resemble the distributed activity that occurs during natural sensations. Because real sensory experiences activate distributed cortical networks — not tiny, localized groups of neurons — the multi-region design mimics more natural patterns of brain activity.

“In the first paper, we used a single micro-LED,” Wu said. “Now we’re using an array of 64 micro-LEDs to control the pattern of cortical activity. The number of patterns we can generate with various combinations of LEDs — frequency, intensity and temporal sequence — is nearly infinite.”

Roughly the size of a postage stamp and thinner than a credit card, the new device also is less invasive. Instead of extending into the brain through a tiny cranial defect, the new soft, flexible device conforms to the surface of the skull and shines light through the bone.

“Red light penetrates tissues quite well,” Kozorovitskiy said. “It reaches deep enough to activate neurons through the skull.”

Successful stimulation

To test the system, the team used mice engineered to have light-responsive cortical neurons. Then, they trained the mice to associate a particular pattern of brain stimulation with a reward. Typically, this task involved visiting a specific port in a chamber.

In a series of trials, the implant delivered a specific pattern across four cortical regions — like tapping a code directly onto neural circuits. The mice quickly learned to recognize this target pattern among dozens of alternatives. Using artificial signals carried by the target pattern, they chose the correct port to receive a reward.

“By consistently selecting the correct port, the animal showed that it received the message,” Wu said. “They can’t use language to tell us what they sense, so they communicate through their behavior.”

Now that the team has shown the brain can interpret patterned stimulation as meaningful signals, they plan to test more complex patterns and explore how many distinct patterns the brain can learn. Future iterations might include more LEDs, narrower spacing between LEDs, larger arrays covering more of the cortex and wavelengths of light that penetrate deeper into the brain.

The study was supported by the Querrey Simpson Institute of Bioelectronics, NINDS/BRAIN Initiative, National Institute of Mental Health, One Mind Nick LeDeit Rising Star Research Award, Kavli Exploration Award, Shaw Family Pioneer Award, Simons Foundation, Alfred P. Sloan Foundation and Christina Enroth-Cugell and David Cugell Fellowship.