Revolutionary wearable device uses light to safely scan the brain

Peering into the human brain has never been easy. For decades, neuroscientists have relied on heavy, expensive machines to measure blood flow and oxygen levels that reveal how the brain works. These tools, while powerful, have been confined to labs and hospitals, limiting access to only a small circle of researchers and patients.

Los Angeles–based neurotechnology company Kernel is poised to change that, shrinking the power of advanced brain imaging into a lightweight, wearable headset.

Kernel Flow, is based on time-domain functional near-infrared spectroscopy, or TD-fNIRS. Unlike older methods that shine a steady stream of light into the head, TD-fNIRS uses ultrafast pulses—on the order of trillionths of a second. By measuring how these pulses scatter and are absorbed, scientists can track oxygen levels in the brain and see how blood flow changes as someone thinks, moves, or learns.

Until now, systems capable of this level of detail have been bulky, costly, and difficult to use. Most clinics and research centers could not justify the investment, leaving TD-fNIRS on the sidelines despite its promise. Kernel Flow addresses this barrier by packaging the technology into a 2.05-kilogram headset that runs from a USB-C cable, making it portable and practical without losing accuracy.

Kernel Flow1 is built from 52 modules arranged in plates that wrap around the head, giving full coverage of the brain’s major regions. Each module functions like a miniature scanner, with two lasers—one emitting light at 690 nanometers and the other at 850 nanometers. These wavelengths are ideal for penetrating human tissue.

Surrounding the lasers are six detectors arranged in a hexagon. Light travels between the lasers and detectors through spring-loaded pipes that maintain steady contact with the scalp, even if the wearer shifts position.

This modular approach solved two problems that have long plagued brain imaging. First, it reduced the overall size of the device. Second, it made maintenance easier, since individual modules can be replaced if needed instead of servicing the entire system. The adjustable headset can also fit different head shapes and sizes, a practical feature for everyday use.

Inside each module, three main parts—lasers, detectors, and optics—work together to collect photons and record the exact time each one arrives. This timing is critical because it allows the system to distinguish between light scattered by deeper brain structures and light reflected from surface tissue. The result is a more precise reading of brain activity.

To prove Kernel Flow1 could perform as well as traditional machines, the team first tested it using “phantoms.” These are physical models that mimic the way human tissue absorbs and scatters light. The researchers applied a set of standardized protocols, including MEDPHOT, which measures how light passes through simulated tissue, and Basic Instrument Performance, which assesses sensitivity and response speed.

The results showed that Kernel Flow1 matched or even surpassed benchtop systems in accuracy. In other words, the compact headset could deliver the same high-quality data as larger, far more expensive equipment.

The next step was to see whether the headset could measure brain activity in real people. For this, researchers turned to a classic neuroscience experiment: finger tapping. When a person taps their finger, blood flow increases in the motor cortex, the brain area responsible for movement.

Two volunteers wore the Kernel Flow1 headset while performing this task. The device captured increases in oxygenated blood and decreases in deoxygenated blood in the motor cortex, producing the same patterns other established imaging systems have documented. In total, the experiment collected data from more than 2,000 channels across the brain.

“Our miniaturized device demonstrated performance similar to benchtop systems, validated by phantom protocols and human neuroscience results,” said Ryan Field, Kernel’s Chief Technology Officer. These exciting results lead to the development of Kernel Flow2.

Kernel Flow 2 isn't just an incremental update—it’s a major leap in hardware and software capability. Similar to the Kernel Flow1 device, Kernel Flow2 is a portable, noninvasive headset that measures brain function using infrared light. By tracking changes in hemoglobin levels and other brain activity signals, it offers a way to study brain function without surgery.

Differential upgrades and features include:

Researchers tested whether its readings remain consistent across repeated sessions, different headsets, and varied conditions. The study involved 49 healthy adults, averaging 44 years old, who attended two sessions each. They completed measurements at rest, while listening to sounds, and during a cognitive control task called the Go/No-Go. This design allowed the team to compare results both within the same day and across separate visits, while also evaluating performance across different Flow2 devices.

At rest, the device produced reliable measurements of hemoglobin concentration, light transmission, slow brainwave fluctuations, and functional connectivity. Only prefrontal connectivity showed weaker reliability, likely due to factors like sleepiness or mood. Even so, further analysis found no systematic drift, confirming stable device performance.

During tasks, the device consistently detected brain activity in expected regions: auditory areas during sound exposure and the right prefrontal cortex during response inhibition. Statistical tests confirmed high reliability, even when switching headsets, and behavioral results were stable across visits. Together, these findings suggest Flow2 can deliver dependable brain activity measurements in both rest and task settings.

Overall, the findings, published in the journal Scientific Reports, show that Flow2 provides reliable, repeatable measurements of brain activity in both rest and task conditions. That matters because for brain-based biomarkers to be useful in real-world clinical settings—such as diagnosing disorders or monitoring treatment—they need to be both dependable and practical. The Flow2 headset’s good performance supports its potential use in future medical and neurological research.

However, the study noted one area needing more exploration: how the device performs across different hair types, since optical devices like this can be affected by hair texture. The researchers suggest future studies should investigate that.

Moreover, while the results are promising for healthy adults, more research will be needed to test the device’s reliability and validity in clinical populations—people with neuropsychiatric or cognitive conditions—to fully unlock its diagnostic and monitoring potential.

One of Kernel Flow’s most notable qualities is how user-friendly it is compared to traditional machines. The spring-loaded light pipes ensure stable contact even when the wearer moves slightly. The headset is powered through a standard USB-C connection, eliminating the need for heavy batteries. At just over two kilograms, it is light enough to wear for extended sessions without discomfort.

The modular design also makes it scalable. Since the device can be manufactured in large numbers at a lower cost, it opens the door for broader use in classrooms, clinics, and even homes. This type of accessibility has been out of reach for TD-fNIRS technology until now.

The development of Kernel Flow marks a major shift. Brain imaging that once required large, specialized labs can now be done with a portable headset that doesn’t compromise on precision.

It allows researchers to measure brain activity continuously, at speeds of up to 200 times per second, capturing subtle and fast changes that could deepen our understanding of cognition and behavior.

Note: The article above provided above by The Brighter Side of News.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.