The technology for real-time 3D ultrasound has existed for years, tucked inside high-end imaging systems at major medical centers. Philips has it. So do Siemens and GE. But the specialized probes that make volumetric imaging possible—those matrix-array transducers that can capture depth alongside width and height—carry price tags roughly ten times what a standard 2D ultrasound system costs.
For a community hospital in rural North Carolina, or a clinic operating on tight margins, that premium keeps 3D capabilities firmly in the "nice to have" category. Which is where things have stayed, more or less, until now.
Lumius Imaging, a Durham startup that emerged from Y Combinator this year, is making a straightforward pitch: What if you could get real-time 3D ultrasound for a fiftieth of the usual probe cost? The company, drawing on research from Duke University's biomedical engineering labs, has built a system around a technology called diffractive acoustic tomography—a method that pairs standard linear probes with what amounts to a carefully engineered slit to generate volumetric images.
It sounds almost too simple. Which is probably why Lumius is careful to ground its claims in peer-reviewed science and acknowledge, prominently, that nothing it's built has FDA clearance yet.
The Slit That Changes Everything
Here's what Lumius calls "True 4D" imaging: real-time 3D with consistent resolution across a 4×4×4 cm field of view. The company isn't relying on expensive matrix arrays. Instead, its diffractive acoustic tomography approach uses off-the-shelf linear probes—the kind already sitting in ultrasound carts across thousands of facilities—combined with a single diffractive element. Think of it as retrofitting existing hardware rather than replacing it entirely.
The system comes with AI-assisted features designed to automatically identify vessels, nerves, and muscle. Lumius has outlined plans for multiple form factors, including plugin modules and full probes, though detailed specifications remain sparse. The company describes the technology as a "3D camera for the body," which is marketing language, sure, but it captures the ambition: making volumetric imaging routine enough that it becomes a default rather than a luxury.
Lumius is starting with vascular access—specifically, central line placement, a common hospital procedure that fails on the first attempt roughly half the time when performed by less experienced clinicians. The company believes 3D guidance could chip away at that failure rate, though proving that in controlled clinical settings will be another matter.
From Duke Labs to Y Combinator
The underlying science came out of Duke's Pratt School of Engineering. On January 29, 2025, a research team that included Lumius's cofounders published findings in Nature Communications detailing the 3D-DAT method. The paper reported near-isotropic 3D imaging using standard linear arrays paired with a diffractive element, with GPU processing enabling reconstruction speeds roughly 50 times faster than earlier photoacoustic approaches.
Duke characterized the breakthrough as transforming standard probes into high-resolution 3D tools via a "tiny slit"—a simplification, perhaps, but one that hints at the elegance of the design. Whether that elegance survives the transition from lab bench to clinical workflow is the question every academic spinout faces.
Lumius joined Y Combinator's Spring 2026 cohort, announcing its acceptance on April 1 and posting its launch video in May. Demo Day arrived in mid-June. The company operates with a lean team of four, according to its Y Combinator profile: CEO Tri Vu, CTO Luca Menozzi, and COO/CFO Chenhang Li, all holding PhDs from Duke in relevant disciplines. Junjie Yao, who led the 3D-DAT research at Duke, serves as scientific advisor and cofounder. Clinical advisors include vascular surgeon Jeffrey Lawson and anesthesiologist Keith Hemmert.
Before Y Combinator, Lumius secured early funding from North Carolina's biotech ecosystem—an NC Biotechnology Center award in February 2025, followed by a Duke-Coulter Translational Partnership grant that August. Not massive sums, but enough to validate the concept and build initial prototypes.
The Regulatory Elephant
Lumius's website carries a conspicuous disclaimer: its statements and products haven't been evaluated by the FDA and aren't intended for medical use. That's standard language for pre-clearance medtech, but it's also a reminder that commercial availability is still somewhere down the road. No FDA submissions have been announced publicly, and the timeline for clearance—especially for a novel imaging modality—can stretch longer than founders typically anticipate.
The company hasn't disclosed pricing details, though the "50 times cheaper" claim suggests a target price point well below the tens of thousands traditional 3D probes command. Whether that cost advantage holds once manufacturing scales and regulatory hurdles are cleared is an open question.
A Crowded Field, A Specific Bet
Lumius isn't entering an empty market. The broader ultrasound sector has been trending toward portability and AI integration for years. Butterfly Network put ultrasound on a chip and turned it into a handheld device. Clarius and Philips Lumify followed with their own pocket-sized systems. Meanwhile, companies like iSono Health have carved out niches with FDA-cleared automated 3D platforms for specific applications—breast imaging, in iSono's case.
What Lumius is wagering on is a middle ground: 3D capabilities at a price point that makes sense for point-of-care settings, somewhere between premium handhelds and full cart-based systems. It's positioning DAT as a platform technology, with a roadmap that extends beyond vascular access into blood clot detection, nerve blocks, tumor diagnosis, and biopsy guidance. Ambitious? Absolutely. But the ultrasound market has shown appetite for technologies that democratize access.
Still, the company is starting narrow—vascular access as the beachhead—which suggests the founders understand the danger of spreading too thin before proving the core use case. Whether diffractive acoustic tomography can do for 3D ultrasound what earlier innovations did for portability—make it accessible enough to become routine—depends on factors well beyond the elegance of the science. Clinical validation, regulatory navigation, manufacturing scalability, and the stubborn realities of hospital purchasing decisions all loom ahead.
For now, Lumius has the published research, the assembled team, and the backing of one of Silicon Valley's most influential accelerators. What it doesn't yet have is a product clinicians can legally use on patients. That gap—between compelling technology and commercial reality—is where most medtech startups either find their footing or quietly fade. We'll know soon enough which path Lumius takes.
