Imagine treating a brain disease with a jab in the arm instead of a hole in the skull. That’s the future a team at MIT is sketching out with a technology they call Circulatronics. It’s a mash-up of tiny microscopic electronics and the body’s own immune cells that can slip through the bloodstream, cross into the brain, and self-install at sites of disease.
Once there, these tiny devices wait for a beam of invisible light from outside the body for instructions. The implants stimulate tiny regions of the brain sometimes just a few dozen microns across.
“Our tiny electronic devices seamlessly integrate with the neurons and co-live and co-exist with the brain cells creating a unique brain-computer symbiosis,” said lead researcher Deblina Sarkar.
Circulatronics
Today, if you need an implant to quiet the tremors of Parkinson’s disease or to probe a seizure disorder, you’re signing up for brain surgery. Surgeons thread metal electrodes deep into the brain through holes drilled in the skull. The approach can transform lives, but it carries all the usual risks that come with cracking your skull open, as well as enormous cost.
Even newer “minimally invasive” ideas, like stent-mounted electrodes deployed through blood vessels, still require a catheter snaked up to the brain. They also can’t easily reach every region or stimulate with sub-millimeter precision.
Deblina Sarkar, who runs the Nano-Cybernetic Biotrek Lab at the MIT Media Lab, wanted something radically different. Instead of pushing big wires into the brain, she and her colleagues asked a question. What if the electronics were so small they could ride along with blood cells, and so smart they’d know where to get off?
Over six years, the team built exactly that.
At the heart of Circulatronics are what the researchers call subcellular-sized wireless electronic devices, or SWEDs. Each SWED is smaller than a blood cell (only a few micrometers across) and only about 200 nanometers thick. Each one is a tiny photovoltaic chip made from organic semiconducting polymers sandwiched between metal layers. When bathed in near-infrared light, they turn that light into electricity, generating a nearly constant voltage of about 0.2 volts and nanoamp-scale currents even at these miniature sizes.
In lab tests, SWEDs continued to generate power even when light had to pass through an entire mouse brain and skull first, producing around half a nanowatt of power at safe light levels. That’s a pitiful amount of power when judged by household standards, but at the scale of neurons, it’s enough to nudge nearby cells into firing.
Sneaking a Chip into the Brain
Electronics that small have a problem, though: the immune system.
Free-floating foreign objects in the blood tend to be gobbled up by white blood cells or filtered out by organs like the spleen and liver. They also have to cross the blood–brain barrier, the brain’s famously picky border control that keeps out most drugs, pathogens, and particles.
Sarkar’s group solved that by turning those white blood cells into accomplices.
They chemically glued SWEDs onto monocytes. These immune cells naturally home in on inflammation and can squeeze through the blood–brain barrier under certain conditions. Using a click-chemistry reaction, they decorated the surface of monocytes with molecular “hooks,” and tagged the SWEDs with matching “loops,” snapping the hybrids together with covalent bonds.
The result is a cell–electronics hybrid: a living courier with a hard-tech backpack. Hence the name Circulatronics, “electronics that circulate through the vasculature,” as the team puts it in their Nature Biotechnology paper.
“Our cell-electronics hybrid fuses the versatility of electronics with the biological transport and biochemical sensing prowess of living cells,” Sarkar said in an MIT news release. “The living cells camouflage the electronics so that they aren’t attacked by the body’s immune system and they can travel seamlessly through the bloodstream. This also enables them to squeeze through the intact blood-brain barrier without the need to invasively open it.”
Casting for Inflammation
During tests of Circulatronics, using mice, the researchers first created a tiny island of inflammation in a deep region called the ventrolateral thalamic nucleus by stereotactically injecting bacterial lipopolysaccharide, or LPS. This is a standard way to provoke an immune response. The injection here is only for the experiment. In a future clinical setting, the inflamed region would come from the disease itself.
Then they injected about 2 million fluorescently tagged cell–electronics hybrids into the animals’ bloodstream via the tail vein. Over the next 72 hours, those hybrids went circulating through the body.
When the scientists later examined brain slices under a confocal microscope, they saw what they had hoped for: clusters of hybrids lodged right in the inflamed thalamic target. In control animals that got SWEDs without the cell carriers, almost none of the devices made it into the brain. In another control group where there was no induced inflammation, the hybrids largely stayed out of the brain tissue.
Chemical analysis revealed that roughly 14,000 SWEDs had self-implanted in the brains of experimental animals. Logistic regression analysis suggested that the presence of the inflamed region was a strong predictor of where the hybrids ended up.
The coolest part about all of this is that the same method can be used to make tiny self-implanting devices across the body, not just in the brain.
“This is a platform technology and may be employed to treat multiple brain diseases and mental illnesses. Also, this technology is not just confined to the brain but could also be extended to other parts of the body in future,” said Sarkar in a press release.
Lighting Up Neurons Without Opening the Skull
Getting chips into the brain is only half the story. The real question is: Do they do anything once they get there?
Before moving to live animals, the researchers sprinkled SWEDs onto cultured neurons in a dish and flashed them with green light. Neurons near the devices started firing in sync with the light pulses, while neurons illuminated without SWEDs stayed quiet.
Then came the in vivo test. After 72 hours — enough time for the hybrids to self-implant in the inflamed thalamus — the team shone a near-infrared (NIR) laser through the intact skulls of some mice. The used 792 nanometer light in 10-millisecond pulses at 20 hertz, for 20 minutes. NIR passes better through tissue than visible light and can be delivered from outside the head.
To see whether neurons responded, the researchers looked for c-Fos, a protein widely used as a marker of recently activated brain cells. In mice whose brains contained self-implanted hybrids and who received the NIR stimulation, c-Fos-positive cells surged to about 318 cells per square millimeter in the target region. In control mice — with hybrids but no light, cells but no SWEDs, or light alone — the numbers were closer to 70–110 cells per square millimeter.
Out of 64 units recorded in mice with hybrids plus NIR light, 14 showed strong, time-locked responses: spikes appearing just after each light pulse, with far more consistent timing than neurons in control animals. Statistical analysis put the timing precision of these spikes in the top 1% compared with light-only and cell-only controls.
Together, the molecular and electrical data make a compelling case that SWEDs are functional, addressable stimulators, activated wirelessly from outside the body, focused to a tiny patch of brain, without any neurosurgeon guiding them into place.
Are the Hybrids Safe Houseguests?
Whenever you hear “brain implant,” your next thought should probably be “No, thanks.” For some patients, however, no is not an option.
In the short term, the team monitored mice that received cell–electronics hybrids for basic health measures: weight, water intake, blood counts, liver and kidney chemistry, and behavior. In open-field tests of locomotion and “novel object recognition” tasks that probe memory, animals with hybrids behaved much like controls.
Using whole-body fluorescence imaging, they tracked the hybrids’ clearance from the body. Within about 10 days, the fluorescent signal dropped back to baseline and ex vivo imaging of major organs suggested the hybrids had been cleared, at least at a detectable level. So it seems to be safe.
When SWEDs were injected directly into the brain (without cells) to study any chronic tissue response, they tended to stick around. Over a six-month period, the number and area occupied by SWEDs in brain tissue stayed relatively constant. Yet histological staining didn’t reveal obvious tissue damage around them.
On one hand, you might want your implants to persist for years to treat chronic diseases. On the other, you’d like options for making them go away once their job is done. In the discussion, the authors suggest future versions could be built from biodegradable materials, or attached to cells with linkers that fall apart over time or in response to a specific cue — like a particular light wavelength, pH environment, or enzyme.
What Does the Future Look Like?
Sarkar has suggested that the technology “holds the potential to make therapeutic brain implants accessible to all by eliminating the need for surgery.” Through an MIT spin-off called Cahira Technologies, she and colleagues aim to move Circulatronics into clinical trials within a few years, with an eye toward conditions like brain cancer, Alzheimer’s disease, chronic pain, and even deadly brainstem tumors that are currently inoperable.
If this sounds exhilarating and unnerving at the same time, that’s because it is. Imagine a world where a teenager with drug-resistant epilepsy, or an older adult with severe depression, could receive a programmable, deeply targeted neural therapy without risking open-skull surgery.
Because SWEDs are controlled by light, stimulation protocols could be adjusted over time — much like current deep-brain stimulators, but without a battery pack under the skin or wires tunneled through the neck. In principle, a single injection could scatter thousands of tiny stimulators across a diseased network, taking on shapes that match each person’s unique anatomy and pathology.
But the unnerving questions shouldn’t be ignored. Who decides where these hybrids are allowed to travel? In the mouse experiments, inflammation is deliberately created in one spot; in a human brain riddled with multiple lesions or smoldering inflammation, the immune system’s roadmap may be more chaotic. What if hybrids lodge in places you didn’t plan for?
A New Approach to Implants
For now, Circulatronics lives in a narrow world: a Nature Biotechnology paper and a lot of excited speculation.
But it also represents a subtle shift in how we imagine our own bodies.
So far, we’ve been used to implants that were rigid, localized things. We have metal rods in spines, titanium in knees, electrodes anchored in brain tissue. You could point to where the implant was. You could take an X-ray and see its silhouette.
The Circulatronics vision is different. It imagines implants as distributed systems: swarms of microscopic devices that ride on cells, spread through tissues, and only become “visible” when they suddenly light up neurons in response to an external signal.
Whether that future ever comes to pass will depend on a mountain of follow-up work: scaling up from mice to larger brains, dialing in safety, handling chronic use, navigating regulatory pathways, and answering hard ethical questions about consent and control.
