In 2026, China is turning 3D printing, or additive manufacturing, into a mass-production industry. A new Ganzhou plant can build up to 50 metal printers a month, and Chinese makers exported 2.46 million consumer printers in the first four months of the year. The shift is now reaching medical 3D printing.

The scale of China’s build-out

This week Shenzhen-based Addireen opened a plant in Ganzhou that can build up to 50 metal printers a month. A week earlier, Eplus3D finished a Beijing factory sized for more than 300 large metal systems a year. China shipped 2.46 million consumer printers abroad in the first four months of 2026, up 44.7%. Metal leader BLT grew revenue about 40% last year, per company figures compiled by Tianxia Gongchang.

Now the West. 3D Systems posted a flat $94.6M quarter. (Healthcare, up 6.9%, carried it.) Stratasys won a patent case against Bambu Lab in court, but no sales ban followed.

Where it reaches medicine

Based on our internal research focusing on keywords related to “3D printing”, “additive manufacturing”, and “bioprinting”, the number of publications in PubMed has also grown significantly from China in the last 10 years (see graph below). The percentage share of articles published has steadily increased to 28%. Surprisingly, the percentage share of US publications has been declining to 17%. One could argue that the quality of the Chinese studies might not meet certain academic standards, and I agree with that skepticism especially given high retraction rate. However, these are numbers and trends to keep an eye on.  

Across the United States, the European Union, and China, the volume of academic medical device research, including hardware, diagnostic hardware, implants, and digital health software successfully reaching commercial channels reveals a steep translation funnel restricted by high capital demands and heavy engineering-to-clinical validation hurdles. In the United States, robust private venture networks and experienced technology transfer ecosystems drive an estimated 10% to 15% of university medical device disclosures to achieve commercial licensing or startup spin-out formation, although fewer than 15% of those entities successfully navigate the multi-tier testing and production scaling needed to achieve final market clearance. The European Union sees a lower medtech translation efficiency, with an estimated 4% to 8% of formal disclosures transitioning into commercial vehicles. While European universities excel in foundational bio-engineering and precision mechanics, early-stage spin-outs frequently hit funding gaps when attempting to navigate the complex pathways to secure a market entry mark, resulting in widespread early acquisitions by multi-national corporations rather than independent scaling. Meanwhile, China has accelerated its medical device and industrial engineering commercialization rate to an estimated 12% to 18%, fueled by an aggressive national pivot toward local manufacturing and high-end hospital hardware independence.

The underlying mechanics of this regional translation are dictated by highly distinct public policies that determine how medical technology is funded and regulated across its lifecycle. In the United States, the Bayh-Dole Act enables academic institutions to securely own and license federally funded intellectual property, creating a decentralized framework where universities work closely with private venture funds. We discussed this topic with UCLA Technology Transfer Office Mark Wisnieski in a recent episode.

However, hardware startups must carefully manage recent policy updates like the CMS substantial clinical improvement mandates, which place strict clinical evidence burdens on breakthrough medical devices seeking public insurance reimbursement. In Europe, the historic emphasis on academic publication metrics over technology transfer has been exacerbated by the structural rollout of the European Medical Device Regulation (MDR), whose increased compliance costs, data requirements, and notified body bottlenecks have significantly delayed spin-out timelines. Conversely, China’s state-directed model bypasses these traditional translation bottlenecks through centralized State Council patent conversion utilization plans and matching grants. These strict policies actively audit and penalize underutilized academic patents, forcing universities to liquidate dormant hardware property directly into industry partnerships while deploying fast-tracked domestic regulatory pathways to prioritize the manufacturing of homegrown medical machinery.

These are manifesting in the medical 3D printing market in these regions, though the data is currently spotty. The US has clearly had a head start in clearing 3D printed medical devices, with 357+ cleared by 2023 and 1900% increase between 2010-2022. Assuming the same rate of growth, cumulative 510(k) cleared devices could reach 1000+ this year. In comparison, according to the same Tianxia data, China had 199 registered 3D-printed medical devices by late 2024, about ten times the 2020 count. It certainly would be worthwhile to closely monitor more granular data from both regulatory bodies moving forward.

Large clinical series are now showing up from Chinese hospitals, like a 224-patient anatomic cage study in late September. That fits the wider trial environment. One 2025 analysis counted 16,612 registered trials in China in 2023 against 9,100 in the US, though that covers every trial type, not just devices, and Chinese registry data is less standardized. Volume follows patient concentration: a US investigator described a Shanghai hospital enrolling 240 patients in two months, a pace he said would take years in the US. Earlier published series on printed cages in this area were small, such as retrospective cohorts of 56 and 60 patients, so a multicenter comparison of 224 is a step up in scale. The regulatory split reinforces the gap. China often requires China-specific trial data for high-risk devices, while FDA accepts foreign data that meets its standards, and most printed spine cages in the US reached the market through 510(k), where only a minority of submissions include clinical data. That is consistent with the publication share above: Chinese hospitals are producing more of the clinical evidence, while the US pathway has not required it. Whether that evidence is rigorous enough, and whether FDA or payers will accept it, is still open.

State support or market logic?

The question underneath all of this: is China’s additive manufacturing (AM) success entirely state support, or is there a real market logic? The state case is strong. A 2017 action plan promised more fiscal support and a revenue target, and since 2018 Beijing has prioritized domestic purchases of machines and powders, per Metal AM. Tianxia Gongchang’s 2026 report lists priority procurement and tax credits of 8–10%. A US congressional commission’s review of Made in China 2025 found that the technologies meeting most targets combined long-term state support, integrated supply chains and scale.

But subsidies do not explain everything. A procurement list cannot account for 2.46 million printers sold abroad in four months, and Bambu Lab, founded in 2020 and backed by IDG Capital, sells to consumers who are free to choose. BLT and Farsoon grew revenue about 40% and 45% in 2025. At TCT Asia in Shanghai, 3DPrint.com reported that exhibitors had stopped selling lasers and build volume and started selling workflow, uptime and use cases.

State support also has gaps: Metal AM noted that lasers and scanners for high-end metal machines were still bought from IPG (US) and Scanlab (Germany). Even in consumer printing, one recent article credits both Chinese policy support and Bambu Lab for the surge in filament demand, without saying how much each contributed. Policy helped create the first buyers and lowered the risk, then the market did the rest, with metal 3D printers probably leaning more on policy and consumer 3D printers on markets. Either way, Chinese metal and consumer printers are steadily gaining ground on Western incumbents.

The constant worry: intellectual property

On Sept. 18 a Texas jury found that Bambu Lab infringed four Stratasys patents and awarded about $27.6M in past damages. Bambu disputes the verdict and plans to appeal, no injunction has been announced, and a Hague court declined in April to block its H2C in Europe, so this is one case, not a verdict on an industry. The clock keeps on running: Stratasys sued in August 2024, and Chinese makers exported millions of consumer printers while the case ran. Courts can protect an inventor, but they move slower than a competitor that scales. The system sometimes works: the European Patent Office upheld Carbon’s dual-cure resin patent on Sept. 24, though that fight did not involve China.

The AI race meets the factory floor

While artificial intelligence remains a central focus of US-China relations, broader technological competition and cooperation are also reshaping the AM industries in both nations. The AI boom is actively influencing the AM sector; for instance, Addireen’s Ganzhou plant focuses on producing pure copper parts like AI-server cold plates, directly generating new demand for printed hardware.

At the shop-floor level, a recent Engineering.com interview highlights how local manufacturers are adapting. Roman Arkhangelskiy, founder of the Boston-area job shop Upside Parts, explains that AI-driven intake and routing have accelerated order-to-print startup times from four minutes down to two or three. Arkhangelskiy notes that his business was established to outperform the three-week turnaround times once common among Chinese suppliers two years ago, warning that the efficiency gap between nations will only widen unless US manufacturing adopts more AI and robotics.

A positive precedent: Meshy

Meshy is perhaps a case for an optimistic read. Founder Yuanming Hu came through Tsinghua and MIT and wrote the open-source Taichi graphics library. The company has roots in both Beijing and Silicon Valley, recently raised about $400M in July at a reported $1.5B valuation and with both Bambu Lab and Creality as customers. One team, two ecosystems, one tool used on both sides of the Pacific. That is the race compounding instead of splitting, and it is the kind of outcome worth protecting.

So, is it a problem?

Depends on whom you ask. Cheap, fast, high-quality hardware is mostly good news for the consumers and hospitals that buy it. The risk sits elsewhere. When machines, materials and clinical volume all scale in one country, the learning curve moves with them. A cleared device is a snapshot. The know-how to make the next one lives with whoever prints the most parts.

The defensible layer for US and European players should not be just the hardware. It is perhaps even more important to focus more on high-value applications, qualified workflows, outcomes data, talent pools, and innovation ecosystems.

A better question we should ask is whether we are investing strategically or still just competing on selling printers.

References

1. 3D Printing Industry. “Addireen opens Ganzhou facility with capacity for up to 50 metal AM systems per month.” Sept. 30, 2026.
2. 3DPrinting Journal. “China Watch #9: PLA makers hit stock market records as 3D printing drives demand, while Eplus3D opens Beijing factory for 300+ large metal printers a year.” Sept. 2026.
3. 3DPrint.com. “AM Asia Watch: China exported 2.46 million 3D printers in four months.” June 11, 2026.
4. Tianxia Gongchang Research. “2026 China Additive Manufacturing and 3D Printing: Market Scale, Competitive Landscape & 5-Year Outlook.” 2026. Accessed Oct. 4, 2026.
5. 3D Printing Industry. “Healthcare Growth Reshapes 3D Systems’ $94.6M Quarter.” Oct. 1, 2026.
6. TCT Magazine. “Stratasys awarded $27.6M in damages in first of two patent infringement cases against Bambu Lab.” Sept. 18, 2026.
7. The Lattice Brief. “PubMed publication share of China and US papers on 3D printing, additive manufacturing and bioprinting (internal analysis, ten-year trend).” 2026.
8. C&EN. “How China seeks to solve its quality control conundrum.” Sept. 2025.
9. 3DHEALS (The Lattice Podcast). “Episode 125: Inside Technology Transfer at UCLA with Mark Wisniewski.” Accessed Oct. 4, 2026.
10. BONEZONE (Dan Cook). “Top Trends in Additive Manufacturing for FDA-cleared Orthopedic Devices.” Sept. 29, 2023.
11. Operative Neurosurgery. “Lateral Mass Fusion Using 3D-Printed Anatomic Cages for Atlantoaxial Dislocation: A Multicenter Comparison With Iliac Crest Autograft.” Sept. 2026.
12. PubMed. “Efficacy of a Lateral Mass Fusion Device Combined with a Three-Dimensional-Printed Model in the Treatment of Craniovertebral Junction Abnormalities.” 2021.
13. Journal of Neurosurgery: Spine. “Comparison of outcomes between 3D-printed porous titanium alloy and polyetheretherketone cages for atlantoaxial intra-articular fusion in craniovertebral malformations.” 2025.
14. MedFlux. “China NMPA Medical Device Registration: NMPA vs FDA Guide (2026).” 2026. Accessed Oct. 4, 2026.
15. Emergent CRO. “Medical Device Clinical Trials: Comprehensive FAQ and Guide.” Accessed Oct. 4, 2026.
16. 3D Printing Industry. “China state Action Plan aims to make 3D printing worth $3 billion by 2020.” 2017.
17. Metal AM. “An inside perspective on China’s thriving metal additive manufacturing industry.” Accessed Oct. 4, 2026.
18. US-China Economic and Security Review Commission. “Made in China 2025: Evaluating China’s Performance.” Nov. 14, 2025.
19. Wikipedia. “Bambu Lab.” Accessed Oct. 4, 2026.
20. 3DPrint.com. “At TCT Asia 2026, China’s AM Industry Looked Ready for Scale: Part 1.” 2026. Accessed Oct. 4, 2026.
21. 3D Printing Industry. “New Update in Stratasys v Bambu Lab Patent Infringement Action: Stratasys Awarded $27.6 Million in Damages, Bambu Lab Disputes.” Sept. 2026.
22. TCT Magazine. “Foundational Carbon dual-cure patent survives challenge from competitor in Europe after appeal.” Sept. 29, 2026.
23. Engineering.com (Ian Wright). “The speed paradox in 3D printing.” Sept. 28, 2026.
24. VoxelMatters. “Meshy 7 review: how can humans keep up?” Sept. 30, 2026.
25. Taichi Graphics. “Project website.” Accessed Oct. 4, 2026.
26. BigGo Finance. “Meshy Closes Nearly $400 Million Series B, Shattering AI 3D Generation Funding Record at Over $1.5 Billion Valuation.” July 2026.
27. China AI Dispatch (Yuzu Xu). “The Third Dimension.” 2026. Accessed Oct. 4, 2026.

Glossary

510(k): The US FDA clearance route in which a device is shown to be substantially equivalent to a legally marketed predicate device. Clinical data are required only in a minority of cases.

Additive manufacturing (AM): Industrial term for 3D printing: parts are built layer by layer from a digital model rather than cut or molded.

Anatomic cage: An implant, typically used in spinal fusion, shaped to match a patient’s anatomy.

Bioprinting: 3D printing with living cells and supporting materials (“bioinks”) to build tissue-like structures.

CAD (computer-aided design): Software files that define a part’s exact geometry. “Dimensionally accurate” means the printed or modeled part matches the design measurements within tolerance.

Cold plate: A metal plate with internal channels that carries liquid coolant past hot chips. Copper is used for its thermal conductivity, and printing can create complex internal channels.

Dual-cure resin: A photopolymer resin hardened in two steps, first by light and then by heat.

European Patent Office (EPO): The body that grants European patents. Its Board of Appeal reviews opposition decisions.

Filament: Plastic wire, such as PLA, fed into desktop extrusion printers.

Frenemy: A rival that also cooperates where interests overlap.

Generative AI (3D): Software that creates 3D models from text prompts or images, as Meshy does.

Injunction: A court order stopping a party from selling or using a product. It is a separate remedy from damages.

Job shop: A contract manufacturer that makes custom parts for other companies.

Made in China 2025: A 2015 national plan to upgrade Chinese manufacturing. The 2017 additive manufacturing action plan formed part of it.

Metal AM: Printing metal parts, often by fusing metal powder with lasers.

Past damages: Money awarded for infringement that has already happened. It does not cover future royalties or sales bans.

Physical AI: Our shorthand for AI that acts on the physical world: robots, automated production cells and the hardware they make.

Series B: A second major venture funding round, usually raised after a company shows product and revenue traction.

Taichi: Open-source programming language for high-performance graphics and physics simulation, created by Meshy’s founder.

Tianxia Gongchang: The Chinese research publisher behind the 2026 AM market report cited here.

Starting draft. Edit freely and upload your version; it will run verbatim.

The post China is printing more of everything. Is that a problem? appeared first on 3DHeals.

DiffractGPT, an open artificial intelligence (AI) model for materials design, has been downloaded roughly 240,000 times on Hugging Face “more than DeepSeek”, told its creator to the room during our recent 3DHEALS virtual event Biomaterials Frontier for Medical 3DPrinting. That creator is Kamal Choudhary, a materials scientist at Johns Hopkins University and a research associate at the National Institute of Standards and Technology (NIST). In fifteen minutes, he laid out what AI can already do in materials discovery: JARVIS, the open materials database he built at NIST; ALIGNN, a graph neural network that predicts material properties; and AtomGPT, a “ChatGPT for materials”. That said, almost none of these can directly benefit bioinks, implantable biomaterials, or medical 3D printing community. This article explains what AI-driven materials design can do today, why biology is the bottleneck, and where the commercial opportunity sits for bioprinting and medical device companies.

What AI can already do

Start with the model behind that download count. DiffractGPT reverses the usual workflow: give it an X-ray diffraction (XRD) pattern and it proposes the atomic structure that would produce it, including for compounds that do not yet exist in any database. Choudhary said at our recent event that it has been downloaded around 240,000 times on Hugging Face, “more than DeepSeek”, he noted. This is a remarkable reach for a tool built for crystallographers. And AtomGPT, his “ChatGPT for materials,” now connects to general chatbots, including OpenAI and Claude, through the Model Context Protocol (MCP), so a language model answers a materials question by querying a physics-grounded tool instead of inventing a plausible-looking wrong one. In a field where a confidently wrong number can send a lab down a six-month dead end, that plumbing matters more than it sounds.

Behind the generative models sits the data, and the scale is easy to underestimate. Choudhary built JARVIS (stands for: Joint Automated Repository for Various Integrated Simulations), an open repository of materials data while at NIST; he says it has around 200,000 users and roughly 100,000 materials. On top of that sits a family of models. ALIGNN, the Atomistic Line Graph Neural Network, was the one to pay attention to. Earlier graph models described a crystal as atoms joined by bonds. ALIGNN added the angles between those bonds, and that single change improved accuracy sharply, by his account up to 44 percent, while staying fast. Its reported error for formation energy is about 0.022 electron-volts per atom, close enough to first-principles calculations to work as a screening tool. His group used it to computationally pick superconductors and a metal-organic framework (MOF) for carbon capture, then synthesized and validated them. Prediction, then a real material. If you want the map rather than the demo, his group’s 2022 review in npj Computational Materials, now cited more than 1,200 times, is the standard reference.

Why biology is still outside

During the event, the moderator Craig Rosenblum asked if the same AI tool can be useful to characterize a material’s “inorganic properties like biocompatibility, the degradation rate, the bioactivity”.

“Great question,” answered Choudhary.

The lack of domain-specific data in biology is where the gap is.

The data problem is lopsided. There is a great deal of it for formation energy, bulk modulus, and band structure, the physical properties of inorganic, crystalline materials. There is very little for the properties that decide whether something belongs in a body, the behavior of a soft hydrogel or an absorbable polymer as it dissolves.

AI has gone deep in certain verticals but left the biological slice mostly untouched.

The loop problem is worse and more structural. Most of these models learn from simulation. To trust them, you check them against experiments, and the experiments are the bottleneck. Characterization tools like X-ray diffraction are slow. They generalize poorly to messy, defect-laden, biological systems. So, the theory-to-experiment loop that would let a model propose a bioink and then learn from how it performed does not close easily. Choudhary was candid that his cleanest results were for near-perfect systems, not for the heterogeneous reality of biomaterials. That said, Choudhary’s own showcase, the AI for Materials Design Laboratory at Hopkins, is a genuine closed-loop facility: robots move samples past automated X-ray, indentation, and laser-impact stations while AI decides what to test next. It is also built for materials in extreme environments, like defense and aerospace, not for biological systems. The machine that would do the same for biomaterials, feeding biological outcomes back into the model, mostly does not exist yet. (See our previous event focusing on AI and 3D printing, where NUS researchers have successfully created a ML-feedback loop to generate bioprinted gum tissue.)

Where the money is

That absence is the opportunity, and Choudhary said.

The winners in AI-driven materials, he argued, will not be whoever trains the biggest general model. They will be whoever owns a proprietary dataset in a specific domain, or builds the narrow, expert model for it. His half-joking example, a “collagen GPT,” is exactly the shape of the bet, knowing one of the speakers is an expert in human collagen. Sell the model no one else can train or sell the data no one else has.

For the 3DHEALS audience, that could be the strategy.

Bioink and biomaterial companies already sit on the scarce asset: batch after batch of biological performance data that never leaves their benches. Instrument the printers. Label the outcomes. Build the loop between what you printed and how the tissue responded. The company that does this for its niche, whether degradable polymers, bioinks, or regenerative materials, will own a moat that a foundation model cannot cross, because the model has never seen the data.

The AI, increasingly, is the easy part. The wet-lab evidence is the hard (and expensive) part, and therefore the valuable one.

References

• Biomaterials Frontier (3DHEALS event): https://3dheals.com/biomaterials-frontier/

• Choudhary et al., “Recent advances and applications of deep learning methods in materials science,” npj Computational Materials (2022): https://www.nature.com/articles/s41524-022-00734-6

• AI for Materials Design Laboratory (AIMD-L), JHU HEMI / CAIMEE: https://hemi.jhu.edu/caimee/center-facilities/aimd-l/

•AtomGPT: https://atomgpt.org   • Choudhary demos (YouTube): https://www.youtube.com/@dr_k_choudhary

•Artificial Intelligence Updates For 3D Printing and Bioprinting (on-demand course)

Organizations

• Hopkins Extreme Materials Institute (HEMI): https://hemi.jhu.edu/

• Center for Artificial Intelligence and Materials Engineering (CAIMEE), Johns Hopkins: https://hemi.jhu.edu/caimee/

• National Institute of Standards and Technology (NIST): https://www.nist.gov/

• NIST JARVIS (Joint Automated Repository for Various Integrated Simulations): https://jarvis.nist.gov/

The post More Downloads Than DeepSeek: AI Model for Materials appeared first on 3DHeals.

Metamaterials could be one of the most consequential ideas in materials science, and 3D printing is what turned them from theory into objects you can hold, implant, and build a company around. This piece explains what they are, how the field has evolved, and why they matter for medicine’s future. Unlike conventional materials like wood or metal, which derive their properties from their atomic structure, metamaterials derive their unique properties from their design at the microscopic or nanoscopic level. In other words, there could, in theory, be infinite kinds of metamaterials, not limited by natural resources but only by human creativity. One recent example of a metamaterial is “auxetic materials” presented by Dr. Jeong Hun Park. (See video below.) Scientists create these materials by arranging small, often repeating structures (like tiny coils or patterns) in a specific way. The precise structure allows metamaterials to manipulate waves, such as light or sound, or mechanical forces in unusual ways.

What is a brief history of metamaterials?

The prefix “meta” means beyond, and metamaterials earn the name by achieving behaviors that go beyond what any natural material can do, not through their chemistry but through the way their internal structure is arranged.

The intellectual roots run back more than a century. Early work on how waves interact with periodic structures dates to around 1904. In 1967, Soviet physicist Victor Veselago worked out the theory of a material with a negative refractive index, a substance that would bend light “the wrong way”, decades before anyone could make one. [5] Roger Walser coined the term “metamaterials” in 1999, and in the early 2000s the first practical, microwave-frequency metamaterials were finally demonstrated in the lab. [7]

What matters for medicine is what happened next. For its first two decades, the field was dominated by electromagnetic metamaterials, the stuff of optical cloaking, superlenses, and antennas. Much of the historical research related to metamaterials is rooted in microwave engineering and antenna beam shaping that emerged after World War II.[17]

Over the past several years, the center of gravity has shifted to mechanical and acoustic metamaterials, where the engineered property isn’t how a material bends light but how it bends, absorbs energy, changes shape, or transmits force. That mechanical branch is the one now reshaping biomaterials and 3D printing.[1]

View this post on Instagram

A post shared by 💡Healthcare 3D Printing💡 (@3dheals)

What are the key features of metamaterials?

Three ideas define a metamaterial:

Structure determines behavior.

In an ordinary material, properties come from its atoms and bonds. In a metamaterial, they come from a designed repeating unit. The geometry of the building block, repeated in a precise lattice, is what produces the property. Change the geometry, and you change the material, without changing the chemistry at all.

Properties that natural materials don’t have.

Classic examples include a negative refractive index (bending waves backward). In the mechanical world, metamaterials with a negative Poisson’s ratio, so-called auxetic materials, get fatter when you stretch them instead of thinner [6]. A negative index also underpins the “superlens”, an optic that could in principle resolve features smaller than the wavelength of light, long thought impossible [16]. Others can have near-zero or even negative stiffness, or absorb specific frequencies of sound or vibration.

Tunability.

Because the property lives in the geometry, it can be dialed in: stiff here, compliant there, gradient in between — across a single part. This programmability makes metamaterials so attractive for engineering the human body, where tissues vary continuously in their mechanical properties.

View this post on Instagram

A post shared by Parametric House (@parametric.3d)

What is the relationship between metamaterials and 3D Printing?

The two are deeply intertwined because a metamaterial’s magic lives in microarchitecture that is often impossible to make any other way. Building thousands of precisely shaped, interconnected unit cells, sometimes at the scale of hundreds of microns, with internal features no drill or mold could reach, is exactly what additive manufacturing does well. In practice, 3D printing is the manufacturing method that made metamaterials real, and metamaterials are one of the most compelling reasons to print rather than machine [3].

That relationship has recently gained a fourth dimension. 4D printing produces metamaterials whose geometry changes over time in response to a stimulus (heat, moisture, light, or even acoustic waves), so a flat 3D-printed sheet can fold into a stent, or a scaffold can stiffen as tissue grows into it [4]. When the printed architecture is designed to interact with living cells, researchers increasingly call these structures meta-biomaterials: architected biomaterials whose geometry itself is the functional element [3].

Metamaterials in healthcare: current applications

This is where the field is moving fastest, and it deserves the spotlight. The common thread is control: the ability to program a material’s mechanical response down to specific regions of a single part is what makes metamaterials so powerful for interfacing with the human body.

A few of the most active areas:

Orthopedic and dental implants.

Solid titanium is far stiffer than bone . Ti-6Al-4V has an elastic modulus around 110 GPa, versus roughly 10–30 GPa for cortical bone and under 2 GPa for trabecular bone [10]. That mismatch causes stress shielding: the implant carries the load the bone should, and the surrounding bone weakens and resorbs. 3D-printed lattice and auxetic metamaterials let engineers lower and tune an implant’s effective stiffness toward that of the host bone, while porous architecture invites bone ingrowth for better fixation [2]. Because the same print run can also match a patient’s individual anatomy, the promise is an implant that is simultaneously stiffness-matched, better tolerated, and less likely to fail over a lifetime of loading.

Tissue-engineering scaffolds.

Architected and auxetic scaffolds don’t just hold cells. Their geometry provides mechanical cues that steer stem-cell differentiation down bone or cartilage pathways, a field known as mechanobiology [9]. Newer work uses acoustically responsive architected biomaterials to guide chondrogenic and osteogenic differentiation, aiming at hard-to-treat conditions like osteoarthritis [12].

Biodegradable implants and scaffolds.

Metamaterial architecture also gives designers control over how a temporary implant disappears. A printed scaffold can support healing tissue and then degrade at a rate set by its geometry as much as its chemistry, so the body gradually takes over the load without a second surgery to remove hardware.

Wearable, in-vitro, and implantable devices.

Mechanical metamaterials are being designed into flexible sensors, conformal wearables, and soft implants [1] — a whole class of “mechanical metamaterials for bioengineering.”

Diagnostics and imaging.

Acoustic and electromagnetic metamaterials can sharpen sensitivity in ultrasound and other sensing modalities, enabling smaller, more precise diagnostic devices. Printed sensors built on metamaterial structures can track physiological parameters with higher sensitivity and specificity than a conventional sensor of the same size [19].

Soft surgical robotics.

Metamaterial structures give soft robots programmable flexibility and shape change, useful for minimally invasive tools that must navigate delicate anatomy [13].

View this post on Instagram

A post shared by 💡Healthcare 3D Printing💡 (@3dheals)

The frontier, and the catch

Metamaterials are a genuine new frontier in materials science, and 3D printing is the key that unlocks them. But turning a striking lattice into an approved medical device is its own challenge: a geometry-defined, load-bearing material has to prove it fatigues and performs predictably [11], batch after batch, in terms a regulator can evaluate. That is the harder, less glamorous work of turning a printed structure into a regulator-legible material system. Much of the real progress over the next few years will happen here. A second reason to get fluent in this now: the narrow palette of qualified printing materials is one of the real constraints on healthcare 3D printing, and metamaterials offer a way around it—designing the property you need rather than waiting for a material that happens to have it [15]. (See our companion piece, “Regulator-Legible Material Systems.”)

If you want to go deeper on where architected meta-biomaterials are heading, this is exactly the terrain of 3DHEALS’ “Biomaterials Frontier for 3D Printing” virtual event — where speaker Ebrahim Yarali (MERLN Institute, Maastricht University) presents his work on auxetic, acoustic-responsive architected biomaterials for skeletal regeneration [12]. Details and free registration [8]: https://3dheals.com/biomaterials-frontier/.

View this post on Instagram

A post shared by 💡Healthcare 3D Printing💡 (@3dheals)

References :

1.  Kazim, M., Pal, A., & Goswami, D. Mechanical Metamaterials for Bioengineering: In Vitro, Wearable, and Implantable Applications. Advanced Engineering Materials, 2025;27(7):2401806. https://advanced.onlinelibrary.wiley.com/doi/10.1002/adem.202401806

2.  Shirzad, M., Zolfagharian, A., Bodaghi, M., & Nam, S.Y. Auxetic metamaterials for bone-implanted medical devices: recent advances and new perspectives. European Journal of Mechanics – A/Solids, 2023;98:104905. https://www.sciencedirect.com/science/article/abs/pii/S0997753822003357

3.  Zadpoor, A.A. Meta-biomaterials. Biomaterials Science, 2020;8(1):18–38. https://pubmed.ncbi.nlm.nih.gov/31626248/

4.  Alanazi, B.N., Ahmed, H.A., Alharbi, N.S., Ebrahim, N.A.A., & Soliman, S.M.A. Exploring 4D printing of smart materials for regenerative medicine applications. RSC Advances, 2025;15(39):32155–32171. https://pmc.ncbi.nlm.nih.gov/articles/PMC12412672/

5.  Veselago, V. The electrodynamics of substances with simultaneously negative values of ε and μ. Soviet Physics Uspekhi, 1968 (orig. 1967). https://iopscience.iop.org/article/10.1070/PU1968v010n04ABEH003699

6.  Kolken, H.M.A., & Zadpoor, A.A. Auxetic mechanical metamaterials. RSC Advances, 2017;7:5111–5129. https://doi.org/10.1039/C6RA27333E

7.  Ziolkowski, R.W. Metamaterials: the early years in the USA. EPJ Applied Metamaterials, 2014. https://epjam.edp-open.org/articles/epjam/full_html/2014/01/epjam140003/epjam140003.html

8.  3DHEALS — Biomaterials Frontier for 3D Printing (virtual event, Sept 10, 2026). https://3dheals.com/biomaterials-frontier/

9.  Yarali, E., Zadpoor, A.A., Staufer, U., Accardo, A., & Mirzaali, M.J. Auxeticity as a Mechanobiological Tool to Create Meta-Biomaterials. ACS Applied Bio Materials, 2023;6(7):2562–2575. https://doi.org/10.1021/acsabm.3c00145

10.  Reimagining Orthopaedic Implants: Mechanobiochemical Innovations to Overcome Stress Shielding (review of implant–bone modulus mismatch and stress shielding). https://www.sciencedirect.com/science/article/pii/S0079642526000290

11.  Zadpoor, A.A. Mechanical performance of additively manufactured meta-biomaterials. Acta Biomaterialia, 2019;85:41–59. https://doi.org/10.1016/j.actbio.2018.12.038

12.  Ebrahim Yarali — researcher profile, MERLN Institute for Technology-Inspired Regenerative Medicine, Maastricht University. https://merlninstitute.com/discover-merln/find-a-merln-member/ebrahim-yarali

13.  Zheng, X., Jiang, Y., Mete, M., Li, J., Watanabe, I., Yamada, T., & Paik, J. Metamaterial robotics. Science Robotics, 2025;10(108):eadx1519. https://doi.org/10.1126/scirobotics.adx1519

14.  Interview with Dr. Jeong Hun Park: Auxetics for Soft Tissue Engineering. 3DHEALS. https://3dheals.com/interview-with-dr-jeong-hun-park-auxetics-for-soft-tissue-engineering/

Related Links:

Regulator-Legible Material Systems: The Real Frontier in 3D-Printed Biomaterials

3D Printing & Biofabrication For Breast Reconstruction (On Demand, 2024)

3DHEALS Biomaterials 2024 (On Demand, 2024)

Interview with Dr. Jeong Hun Park: Auxetics For Soft Tissue Engineering

Interview with Julien Payen: Lattice Medical
Interview Dr. Mohit Chhaya: BellaSeno
Interview with Esther Valliant: Bioglass for 3D Printing

The post Metamaterial and 3D Printing appeared first on 3DHeals.

For most of the past decade, the story of 3D printing in medicine was a story about shape. Patient-specific anatomy, impossible geometries, surgical guides tailored to a single body, frequently touted as “complexity for free”. The printer’s magic was that it could make forms that would be impossible or very expensive to make by mold injection. In 2026, the center of gravity has moved. The defining question is no longer “What shape can we print?” but rather “What material can we print, and will a regulator ever accept it?” That shift has a name worth adopting: the move toward regulator-legible material systems.

What “legible” actually means

A regulator does not approve a material just because it is “cool”. A regulator approves evidence, or documentation, that a material behaves the same way every time, degrades on a predictable timeline, releases nothing harmful, and can be manufactured to a specification you can point to. If a material cannot be described in those terms, it is effectively illegible to an agency like the FDA or a European notified body operating under the EU MDR. They have no framework within which to say yes. “Legibility,” then, is the property of being readable by the people who decide whether something can be used in a human body.

For years, printed biomaterials sat squarely in the illegible zone. A laboratory could print a stunning scaffold, but the material was a one-off: no agreed test methods, batch-to-batch variability, a degradation profile no one had fully characterized, and mechanical properties that shifted with every change of machine or setting. It was brilliant science that a reviewer had no way to sign off on, and just as importantly, no way for a company to build a product or an investor to underwrite a business around.

Why 3D printable biomaterials are uniquely hard to read

The core difficulty is that in additive manufacturing, the material and the process are inseparable. A conventional implant polymer arrives with a datasheet; you can reason about it in isolation. A photopolymer resin cured layer-by-layer on a specific printer, at a specific wavelength, with a specific post-cure and wash step, is a different material depending on how it was made. Residual monomer, degree of cure, layer adhesion, porosity, and even sterilization can each move the biological and mechanical outcome. The FDA’s 2017 guidance, Technical Considerations for Additive Manufactured Medical Devices, makes exactly this point: it asks manufacturers to control and document the entire workflow — design, build parameters, post-processing, cleaning, and testing, because the finished device’s safety is a function of the whole chain, not the starting resin alone.

This is why the field has begun talking about material systems rather than just materials. A material system is the full package that makes a printed biomaterial usable and clearable: the feedstock chemistry plus the validated print parameters, the post-processing and sterilization steps, and the standardized test methods that prove the finished part performs. Change one link, and you may have changed the material; the system is the unit that must be defined, controlled, and, importantly, reproduced.

The machinery of legibility: standards

Legibility is manufactured by standards. They are the shared language that turns a private lab result into something an outside reviewer can trust. Several bodies of work matter here. ISO/ASTM 52900 and the broader 52900 series provide additive manufacturing with common terminology and process definitions. ASTM Committee F42 develops additive manufacturing standards, while ASTM Committee F04 governs medical and surgical materials and devices, including the absorbable polymers at the heart of many printed scaffolds. ISO 10993 defines how the biological safety of a material is evaluated (cytotoxicity, sensitization, and the rest), and quality-system standards such as ISO 13485 wrap the whole operation in documented process control. When a new absorbable polymer gains an agreed test method through F04/F42, it crosses a threshold: it stops being “interesting research” and becomes something a regulator can read.

Legibility across the three material tracks

The push toward legibility is playing out simultaneously across the field’s three frontiers. In the soft track, function-first bioresins and degradable photopolymers, the challenge is characterizing the degradation and drug-release behavior of materials that are, by design, changing over time. In hard-track ion-releasing ceramics and bioactive glass, legibility means controlling surface chemistry and post-processing to ensure that a printed or coated implant integrates with bone predictably. In the structural track, titanium and architected lattices mean proving that a geometry-defined, load-bearing material fatigues and performs as its design promises. Each track has the same underlying task: converting a novel material into a documented, reproducible, testable system.

Why this is an investment thesis, not just a regulatory chore

The commercial consequence is the part founders and investors should internalize. The moment a class of printed biomaterial becomes regulator-legible, it crosses from research risk into a fundable, buildable product. Legibility is what lets a material become the “investment-grade core” of a device rather than a promising figure in a paper. As the healthcare 3D-printing market races toward roughly $33 billion by 2031, the binding constraint is shifting from the printer to the material and the surrounding system. The companies that win will be the ones that make their materials legible fastest.

Where this conversation is happening

These are exactly the questions on the table at 3DHEALS’ “Biomaterials Frontier for 3D Printing” virtual event (September 10, 2026), which convenes the researchers, founders, and industry leaders defining the shift — including a bioresorbable-polymer and ASTM standards leader from Poly-Med, a biofabrication founder scaling human collagen (Arrakis Bio), an AI-driven materials-discovery pioneer from Johns Hopkins, a 4D-materials founder, an architected-meta-biomaterials researcher, and a calcium-phosphate bioceramics veteran in the moderator’s chair. If the frontier of 3D-printed medicine is the move from printing structure to printing function, and from lab novelty to regulator-legible systems, then this is a session built around the people making that move real. Details and free registration: https://3dheals.com/biomaterials-frontier/.

References

1.  U.S. Food & Drug Administration. Technical Considerations for Additive Manufactured Medical Devices — Guidance for Industry and FDA Staff. December 2017. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/technical-considerations-additive-manufactured-medical-devices

2.  ISO/ASTM 52900:2021. Additive manufacturing — General principles — Fundamentals and vocabulary. https://www.iso.org/standard/74514.html

3.  ASTM International, Committee F42 on Additive Manufacturing Technologies. https://www.astm.org/committee-f42

4.  ASTM International, Committee F04 on Medical and Surgical Materials and Devices. https://www.astm.org/committee-f04

5.  ISO 10993-1:2018. Biological evaluation of medical devices — Part 1. https://www.iso.org/standard/68936.html

6.  ISO 13485:2016. Medical devices — Quality management systems — Requirements for regulatory purposes. https://www.iso.org/standard/59752.html

7.  European Union. Regulation (EU) 2017/745 on medical devices (MDR). https://eur-lex.europa.eu/eli/reg/2017/745/oj

8.  Mordor Intelligence. Healthcare 3D Printing Market — Size, Share & Forecast (2026–2031). https://www.mordorintelligence.com/industry-reports/global-3d-printing-market-in-healthcare-industry-industry

9.  3DHEALS. Biomaterials Frontier for 3D Printing (virtual event, Sept 10, 2026). https://3dheals.com/biomaterials-frontier/

The post Regulator-Legible Material Systems: The Real Frontier in 3D-Printed Biomaterials appeared first on 3DHeals.

A biodegradable, 3D-printed pigtail stent for gastric leaks can be a valuable teaching tool for medical 3D-printed device entrepreneurs. This mental exercise with a recent novel design shows the reimbursement logic every device team should learn before they fall too deep in love with their own inventions. Medical device reimbursement in the United States often rewards factors other than clinical performance. A recent preprint makes the gap easy to see. It describes BRIDGE, a 3D-printed, biodegradable double-pigtail stent for draining gastric leaks after sleeve gastrectomy. The design is clever. Its lattice mid-section is built from a triply periodic minimal surface (TPMS). It bends around a radius that is roughly seven times tighter than that of a commercial biliary stent, without kinking. It drains about twice the fluid. And because it is printed from a biodegradable resin, it is designed to dissolve on its own rather than be removed during a second endoscopy.¹

So it is better.

However, the question that decides whether it becomes a product is a different one. What gets paid, and to whom? In the United States, clinical superiority and payment are only loosely related. The space between them is where most device companies get their education. An endoscopic leak drain is a good place to learn the lesson. Here are some of the steps to reason it out.

What the procedure pays today

Start with the problem the device treats, because the money follows the problem. Gastric leak occurs in 1% to 3% of primary sleeve gastrectomies and as many as 10% of revisions. It is the complication surgeons fear most. It turns a routine bariatric case into weeks of drainage, repeat procedures, and sometimes an intensive-care stay. The mainstay treatment is endoscopic internal drainage with double-pigtail stents. It closes most leaks. And it does so with stents designed for the bile duct and pressed into service off-label because nothing better is available.¹

Here is the first thing to absorb. Medicare pays the physician for the procedure, not for the stent.

And the procedure it covers depends on how the pigtail is placed. If an interventional radiologist places it through the skin into the collection, the claim is a percutaneous drainage code with its own predictable payment. If an endoscopist deploys it transluminally across the leak, the claim is coded as an upper-GI endoscopy code. That is the route BRIDGE is built for, and the coding gets murkier. The closest listed option, 43240, is written for draining a pseudocyst.² A post-surgical leak is not a pseudocyst. So many of these cases land on 43499, the “unlisted” upper-GI code a payer reviews and prices by hand.³ That invites delay and denial.

Either way, the stent itself is a supply. Its cost is buried inside the facility payment for the procedure. That is an outpatient APC. Or, for the septic inpatients, most of these are DRGs. A percutaneous drain runs a hospital somewhere between $28 and $55 apiece.⁶ A fancier printed device might cost more to make. But the hospital’s payment does not move to cover it. Build a better version of a bundled supply, and it inherits the price of the thing it replaced.

(See Glossary section for all abbreviations.)

What actually changes the payment

If a superior device does not automatically earn more, what does? Three things, from the flimsiest to the most durable.

The first is a temporary top-up.

Medicare will pay a little extra for a while. Outpatient, this comes through pass-through. Inpatient, it comes through the New Technology Add-on Payment (NTAP). The price of entry is the same for both. The device has to be new. It has to be expensive enough that the existing bundle can’t absorb it. And the hard part is that it has to be a substantial clinical improvement over what’s already in use.⁴ There is a gate hidden inside the word “new.” CMS also asks whether the device is already adequately described by an existing payment category. This is where a genuinely new design earns its keep. A better plastic pigtail looks like every other pigtail and gets waved into the bundle, but a device that dissolves in the body is harder to call the same thing you already pay for. Even after clearing the bar, the money is just temporary. Pass-through runs up to three years. Then the device drops back into the bundle.⁵

The second is a dedicated device code. A new CPT code is the more durable move. But it follows a new service, not a better instrument. That is the quiet opportunity here. Endoscopic internal drainage of a leak has no code that actually fits it. That is why the field is stuck between a pseudocyst analogy and an unlisted code. A device built specifically for the procedure gives the device maker something to hang a Category III code on. That is the emerging-technology tier. It records that the procedure is happening. With utilization and outcomes data behind it, that code can grow into a paying Category I code. The process is slow, measured in years. Like it or not, it will help the competitors too. It is the only lever that fixes billing for the whole field rather than renting a payment for one product.

The third isn’t a billing lever. For a device like this, it is often the one that matters most. A leak admission is expensive and long. Under a fixed DRG, the hospital swallows every extra day and every repeat scope. A device that empties the cavity faster, keeps the patient off a second or third procedure, or shortens the septic stay puts real money back in the hospital’s pocket. No code change required, but value is added. This lever sells on a spreadsheet the hospital keeps.

Which features carry weight

Sorting BRIDGE’s advantages against the clinical-improvement test is a useful exercise, but it doesn’t weigh them equally.

Biodegradability can be a strong one. Today, a patient undergoes roughly three endoscopies over about eight weeks. A stent designed to dissolve in six to eight weeks eliminates the need for a retrieval trip entirely.1 “One fewer procedure per patient” is a true improvement, and it is easy to measure.

That said, there is a potential wrinkle to the economics. The retrieval the new device deletes is a procedure the endoscopist currently bills. Innovation improves the system’s economics, partly by reducing a provider’s revenue. That is exactly why early adopters will be integrated systems and bundled-payment programs, not fee-for-service proceduralists. (Remember what Charlie Munger’s theory behind “incentives” is.) Less migration comes next. Every stent that stays put is a retrieval, and a failed-drainage rescue that the hospital doesn’t have to do. Better flow and easier placement are real but soft. “Easier for the endoscopist” isn’t something Medicare rewards unless it results in a shorter procedure or fewer complications that can be easily measured.

There is a catch a reviewer will find first and can sink the thesis. Once a device can’t be removed, its degradation timing is its safety profile. Dissolve too early, and it fails while the leak is still open. Linger too long, and it becomes the source of granulation, obstruction, or a fragment that drifts somewhere you can no longer reach it. So the elegant “no removal” story and the “less migration” story have to be proven together. The evidence that carries weight is the degradation curve’s reliability. Not just its average, but how much it varies across real patients. That is where the clinical dollars go. It is also where a printed polymer either earns trust or loses it. The industry has learned expensive lessons on resorbable stents very recently. 8

The 3D printing angle

The 3D printing that enables the lattice and patient-specific fit does not qualify for a reimbursement category. Nobody pays more because a device was printed. What changes with 3D printing is the regulatory road. A biodegradable, architected, possibly patient-matched implant probably won’t slide through on a 510(k) predicate. It is more likely to face a de novo or PMA review. That costs time and money. And, awkwardly, it also strengthens the payment case. Devices that clear the more rigorous regulatory pathways have historically fared better when they request a separate payment from CMS. The higher bar and the better lane turn out to be the same door.

Conclusion

For a device like this, here are some potential strategies to think about. Sell to hospitals on avoided cost now, while collecting the migration, reintervention, and length-of-stay numbers. Take those numbers into a pass-through or add-on application next. Aim at a moving target. CMS has proposed requiring even breakthrough-designated devices to demonstrate substantial clinical improvement again, starting with 2028 applications.7. Finally, go earn a real code. It is the only step that changes billing for good rather than borrowing it.

The lesson under all of it: better is the price of admission, not the final payout. In US device reimbursement, the money follows evidence of a difference a patient and a payer can both feel.

Which lever is your particular improvement actually pulling?

This piece is an educational illustration of US reimbursement concepts for new medical devices, not legal, coding, or reimbursement advice. Codes, rates, and CMS policies change; confirm current figures against primary CMS sources before relying on them.

Glossary

• 510(k) FDA clearance route for a device shown to be substantially equivalent to an existing (predicate) device. The fastest, lowest-burden pathway.
• Additive manufacturing / 3D printing: Building a part layer by layer. Enables lattice geometries and patient-specific shapes that extrusion can’t produce.
• APC (Ambulatory Payment Classification): The bundled unit Medicare uses to pay hospitals for outpatient procedures. Devices used are packaged into it.
• Bundled supply: A device whose cost sits inside a broader procedure payment instead of being reimbursed on its own.
• Category I / Category III CPT codes: Category I codes are established procedure codes with set payments. Category III codes are temporary for emerging services that track usage before potentially becoming Category I.
• CMS (Centers for Medicare & Medicaid Services): The federal agency that sets Medicare payment rules and rates.
• CPT (Current Procedural Terminology): The AMA code set that identifies procedures for billing. The code, not the device, usually drives payment.
• De novo FDA: pathway for novel low-to-moderate-risk devices with no predicate, creating a new device classification.
• DPS (double-pigtail stent): A drainage stent with curled ends that anchor it. Used off-label for endoscopic internal drainage of leaks.
• DRG (Diagnosis-Related Group): The bundled unit for an inpatient stay. The hospital receives a fixed amount regardless of individual supply costs.
• EGD (esophagogastroduodenoscopy): Upper-GI endoscopy. The procedure family under which transluminal leak drainage is coded.
• EID (endoscopic internal drainage): Placing a pigtail across a leak so one end sits in the cavity and the other in the GI lumen, draining internally.
• FDA (Food and Drug Administration): The agency that authorizes devices for marketing. Clearance is separate from payment.
• HCPCS Level II (C-code): Codes that can identify a specific device for tracking or pass-through. They do not by themselves add payment.
• NTAP (New Technology Add-on Payment): A temporary inpatient add-on above the DRG for qualifying new technologies.
• Pass-through (Transitional Pass-Through, TPT): A temporary outpatient add-on (up to three years) for qualifying new devices before they are folded into the APC.
• PMA (Premarket Approval): The most rigorous FDA pathway for high-risk devices. Requires clinical evidence of safety and effectiveness.
• SCI (substantial clinical improvement): The CMS test asking whether a technology meaningfully improves diagnosis or treatment over existing options. The gate for add-on payments.
• TPMS (triply periodic minimal surface): A class of mathematically defined lattice geometries (for example, the Gyroid) used here to tune a stent’s flexibility and flow.

References

1. Phowarasoontorn P, Ko Y, Makhambetova Z, et al. Biodegradable Architected Stents for Endoscopic Internal Drainage. bioRxiv; posted May 12, 2026. doi:10.64898/2026.05.08.723751. Preprint. Not peer reviewed.
2. American Society for Gastrointestinal Endoscopy / American Medical Association. CPT code 43240: Esophagogastroduodenoscopy, flexible, transoral; with transmural drainage of pseudocyst (includes placement of transmural drainage catheter[s]/stent[s], when performed, and endoscopic ultrasound, when performed). ASGE EGD Coding Sheet. asge.org
3. American Gastroenterological Association. Coding Corner (Clinical Gastroenterology and Hepatology): where no specific CPT code exists for an endoscopic procedure, report the unlisted code 43499 and verify coverage with the carrier in advance. cghjournal.org
4. Centers for Medicare & Medicaid Services. New Medical Services and New Technologies (NTAP eligibility: new, costly, and a substantial clinical improvement over existing technologies). cms.gov
5. Gettysburg Healthcare Consulting. Medicare Transitional Pass-Through Payment (criteria; up to three-year duration before bundling). policypros.net
6. IndexBox. Percutaneous Drainage Catheters Price in the United States, Market Insights (2026). indexbox.io
7. Kelly S. CMS proposes repeal of add-on payment path for breakthrough devices. Healthcare Dive / MedTech Dive; April 17, 2026. healthcaredive.com
8. https://www.tctmd.com/news/fda-warns-risk-major-adverse-cardiac-events-absorb-bvs

The post Better Isn’t Billable: How a New (3D-Printed) Medical Device Actually Gets Reimbursed appeared first on 3DHeals.

Physical AI is artificial intelligence that can sense, reason, and act in the physical world through robots, devices, instruments, and automated systems. In healthcare, that means AI is no longer limited to software that analyzes records or images; it is also showing up in surgical systems, rehabilitation platforms, wearables, smart medical devices, hospital robots, and automated laboratory infrastructure. There are several motivations behind writing this guide: 1. The increasing number of Pitch3D startups is now deploying physical AI to build or defend their existing products. 2. An increasing number of incumbent AI giants, including leading companies like NVIDIA, or venture capital firms, believe the next phase of AI-driven growth will be from this category of technologies. In part one of this guide, we focus on general concepts and players in current healthcare and life science sectors.

What Physical AI Is and Is Not?

While most well-known examples of physical AI applications are autonomous vehicles and robots, digging a little deeper into this new concept will expand the reader’s visual field to fields far beyond these two examples, and perhaps inspire innovation.

There are many definitions out there, but here is one I have concluded based on several references:

Physical AI is an integrated software-and-hardware system that can perceive and understand the real world (think objects, spatial relationships, dynamics, constraints, cause-and-effect, etc.). It then uses reasoning to adapt actions that would complete tasks and interact with the real world.

Perception. Interpretation. Action.

These three words sum up the core of physical AI. However, systems that rely primarily on fixed rules, task-specific processes, or manually engineered world models do not qualify as physical AI according to IQT [1].

Many industrial robots, such as those in an automobile production line, are in this category. True physical AI hardware or software does not rely on pre-defined signals or programming.

Why Physical AI matters in healthcare?

Healthcare is full of physical workflows: clinicians perform procedures, patients interact with devices, laboratories handle samples, and manufacturers run tightly controlled production systems. Physical AI matters because it brings intelligence into real-world processes by combining sensing, decision-making, and action, often under strict constraints on safety, reliability, timing, and regulatory requirements in healthcare systems.

This shift is important because many of the biggest opportunities in medicine are not purely digital. Globally, healthcare is in crisis. Here in the U.S., this crisis is driven by an increasing gap between supply and demand due to a healthcare labor shortage and an exponential rise in healthcare costs. Many believe that better technologies are our salvation to solve this crisis. However, better care often depends on how well these new technologies are delivering. For example, how well a robot assists a surgeon, how accurately a wearable detects a change in physiology, how safely an exoskeleton adapts to a patient’s gait, or how efficiently a lab automation system handles samples and experiments. Many now believe and actively invest in physical AI, hoping this will unlock the potential of existing healthcare technologies.

What are the major categories of physical AI?

A practical way to understand physical AI is to group it by the type of system involved.

View this post on Instagram

A post shared by 💡Healthcare 3D Printing💡 (@3dheals)

Embodied robotics

Embodied robotics includes systems that move and manipulate the world, such as surgical robots, rehabilitation robots, exoskeletons, and other intelligent machines that use sensors and AI models to perceive environments and execute tasks. In healthcare, this is the most visible facet of physical AI because these systems directly support procedures, therapy, and mobility. They interact with patients and providers directly.

Robotic surgical platforms are one of the clearest examples. These systems combine imaging, motion control, sensing, and AI-assisted guidance to improve precision, dexterity, and workflow during minimally invasive procedures.  

Many would agree that Intuitive Surgical has a near monopoly in this sector. However, this landscape is changing rapidly with more flexible and cheaper systems like CMR Surgical and Moon Surgical now available to financially constrained healthcare buyers. Many major medical device companies are investing heavily in robotics, ranging from internal initiatives to M&A. Almost all orthopedic device companies now own surgical robotics divisions and surgical planning software as part of the device offering.

Autonomous mobile systems

This category covers robots and autonomous platforms that navigate physical spaces, avoid obstacles, and complete transport or inspection tasks. In hospitals and health systems, examples include logistics robots that move medications, meals, linens, or specimens between departments. Swisslog Healthcare, Diligent Robotics, Ottonomy, and Aethon are just a few examples in this operational layer of healthcare automation.

Sensor-driven smart devices

Many physical AI systems in healthcare are not robots at all. They are devices that sense physiological signals, interpret them, and adjust therapy or guidance in real time. Examples include wearables, implants, monitoring tools, and therapeutic devices that continuously interpret physiological or environmental data and adjust behavior accordingly. This is one of the most important categories in healthcare because many high-value products, from closed-loop insulin systems to connected respiratory devices, sit at the intersection of embedded sensing and real-time AI. Medtronic, GE Healthcare, Siemens Healthineers, Abbott, Philips, ResMed, and Omron illustrate the range of companies embedding AI into imaging systems, chronic disease management, remote monitoring, and home care devices. Some consumer-based wearables, like the Oura ring and the Eight Sleep, are also popular smart devices.

Industrial and lab automation

Physical AI also includes intelligent automation in laboratories and biomanufacturing settings, where robotic handlers, liquid-handling systems, imaging stations, and quality-control systems coordinate physical workflows. In life sciences, this category matters because scientific progress increasingly depends on automating repetitive experimental tasks while preserving precision and traceability. That said, many have failed to achieve this “automation of biology” in the past decades, and it is worthwhile to pause and reflect upon this. One major reason for such failure is perhaps due to trying to use deterministic engineering systems to interact with a noisy, context-dependent, dynamic, and complex biological system. Biology would require physical AI to achieve real-time perception, interpretation of complex systems and signals, and adaptation.

Some of the additional bottlenecks that are open to innovations also include:

• Tacit and Unspoken Knowledge: A significant source of failure lies in the disconnect between biologists and automation engineers. Unwritten, “feel-based” bench protocols (like precise mixing or cell handling) are incredibly difficult to translate into rigid code. (An LLM could be useful to enabling scientists and AI to collaborate to optimize protocols.)
• The “Black Box” of Biological Data: Generating more data through automation doesn’t equate to understanding if the data analysis pipelines are not equally advanced. Biology lacks the unified, open-source ecosystems (like Python/GitHub) found in traditional software development, leaving methods fragmented.
• Lack of Mass Standardization: Unlike next-generation sequencing, which has collapsed in cost due to miniaturization, many general cellular assays remain highly expensive and rigid, depriving automated pipelines of affordable inputs.

Thermo Fisher Scientific, Danaher, Agilent, Tecan, Hamilton, Sartorius, Cytiva, Azenta, HighRes Biosolutions, and Beckman Colter are among the companies building the hardware, robotics, and orchestration layers that turn laboratories and production environments into more intelligent physical systems.

Rehabilitation robots and assistive devices

Rehabilitation and assistive technologies form a distinct category because they combine sensing, adaptation, and direct interaction with the human body. These systems include gait trainers, neurorehabilitation robots, exoskeletons, smart prosthetics, and adaptive orthotics that tailor support to the user’s movement and condition. Pysonic, a San Diego-based company, offers highly durable, touch-sensing robotic arm prosthetics to both amputees and large industrial manufacturers. Companies such as Ekso Bionics, ReWalk Robotics, and Fourier Intelligence illustrate how physical AI can directly influence recovery, training intensity, and day-to-day mobility support.

Where Pitch3D startups fit?

The 3DHEALS Pitch3D program focuses on fundraising in healthcare 3D printing and bioprinting, but its scope also includes adjacent technologies such as AI and machine learning, robotics, workflow software, 3D scanning, VR and AR, and data-driven planning tools. That makes Pitch3D a useful lens for understanding how physical AI overlaps with healthcare 3D technologies.

Not every Pitch3D company should be labeled a physical AI startup. Some are primarily materials, printing, or manufacturing businesses. A practical test is whether the company connects intelligent perception or planning to a real-world physical intervention, such as a surgical workflow, a patient-specific implant, a smart device, or an automated production process.

Some of the recent Pitch3D startups stand out as physical-AI-adjacent because they connect digital models to physical care delivery:

Psyonic– uses physical AI by combining high‑speed motor control, multi‑touch sensing, and intelligent signal interpretation in its Ability Hand, creating a bionic prosthesis that can feel, adapt, and perform dexterous physical actions for both humans and robots.

Carlsmed– uses physical AI to turn imaging and outcomes data into personalized 3D‑planned spinal surgeries and patient‑specific implants that precisely match each patient’s anatomy and alignment.

Cosm – uses physical AI by combining pelvic ultrasound imaging, AI‑driven cloud software, and 3D printing to design and manufacture personalized pessary devices that precisely match each patient’s anatomy and pelvic floor dynamics.

Vent Creativity– uses physical AI by creating AI‑powered digital‑twin bone and joint models that simulate forces and motion, enabling FDA‑cleared 3D surgical planning tools that directly inform implant design and surgical execution

Zylo3D – uses physical AI in digital dentistry by combining advanced 3D printing hardware with intelligent workflows that let dental offices produce patient‑specific prosthetics and devices on‑site, tightly coupling design and fabrication to real mouths

These companies are especially interesting because they show that physical AI in healthcare is not limited to humanoid robots or autonomous hospital machines. It also includes systems that transform imaging, planning, and patient-specific data into physical devices, surgical tools, implants, and treatment workflows.

Challenges that define this market

The concepts of AI and robotics are decades-old and century-old ideas, respectively. However, while current AI enthusiasm is mainly driven by advances in large-language models, achieving physical AI will be even more challenging and expensive. Hence, why Jensen Huang is so excited about physical AI as the next frontier, anticipating an even larger-scale economic windfall if society truly buys into this vision.

Physical AI in healthcare faces an even higher bar than consumer or enterprise AI because it operates in environments where errors can directly affect patient outcomes. Negative patient outcomes from AI operations in healthcare systems may create an even higher psychological barrier to future adoption in an already conservative industry.  Safety, validation, human oversight, interoperability, latency, and regulatory compliance are therefore central design requirements rather than afterthoughts.

Another challenge is that many healthcare startups still operate in narrow workflow slices. A company may have strong AI for imaging or design but limited physical automation, or strong hardware with limited intelligence. The most durable category leaders will likely be those that integrate perception, control, workflow software, and regulatory execution into coherent end-to-end systems.

What to watch next?

The next phase of physical AI in healthcare will likely be shaped by three converging trends. First, robots and devices will become more adaptive and agentic; second, labs and factories will become more software-orchestrated; and third, 3D design, simulation, and world models will become more central to medical product development and procedural planning.

For healthcare innovators, investors, and startup founders, the most important question is no longer whether AI belongs in medicine. The better question is where intelligence should live in the physical workflow: in the robot, in the implant design process, in the smart device, in the rehabilitation platform, or in the automated lab system that enables the next breakthrough. It seems that we are only limited by our imagination.  

References:

1. https://www.iqt.org/library/what-is-physical-ai-a-definition-and-framework    
2. https://www.ibm.com/think/topics/physical-ai  
3. https://www.synopsys.com/glossary/what-is-physical-ai.html   
4. https://www.hpe.com/us/en/what-is/physical-ai.html  
5. https://www.medtechdive.com/news/fda-ai-medical-devices-growth/728975/ 
6. https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-and-Global-Robotics-Leaders-Take-Physical-AI-to-the-Real-World/ 
7. https://blogs.nvidia.com/blog/ai-medical-devices-gtc-2024/
8. https://intuitionlabs.ai/articles/top-20-medtech-companies-using-ai-2025
9. https://3dheals.com/pitch3d/     
10. https://pmc.ncbi.nlm.nih.gov/articles/PMC10390055/

The post What is Physical AI? – A Guide for Healthcare appeared first on 3DHeals.

For years, “print a crown while the patient waits” has been more aspiration than clinical routine. Milling has dominated same-day restorative dentistry, and resin-based 3D printing has largely been confined to models, surgical guides, and temporaries. That balance appears to be shifting in light of the latest news in dental 3D printing. Across the first half of 2026, a cluster of regulatory clearances, material launches, and academic results suggests that permanent, chairside-printed restorations are moving from novelty to a real clinical option. Why is this such a big deal? The current permanent crown workflow involves at least two office visits and a minimum of two weeks of waiting time for the crown to be ready for placement. During the weeks between visits, patients often have to wear a clunky, essentially nonfunctional temporary crown. With chairside permanent crown availability, patients can potentially receive same-day treatment, avoiding additional trips to the dentist and the inconveniences of a temporary crown. So it is a huge deal. However, after some investigation, there appear to be real caveats that any dentist or patient should understand before assuming printed crowns can simply replace milled or sintered ones.

The Headline Development: SprintRay’s FDA Clearance for Porcelain Crowns

The most consequential news is SprintRay’s FDA clearance to 3D print porcelain crowns. The company’s CEO, Amir Mansouri, has framed this as roughly 10 times the addressable market of its existing nightguard business, since SprintRay estimates that around 40 million crowns are produced annually in the United States, almost all of them currently routed through outside dental laboratories with a multi-week turnaround. SprintRay’s pitch is that its printers can now produce a crown chairside in 10 to 20minutes, collapsing weeks of lab logistics into a single appointment.

What makes this more interesting is the pricing strategy attached to it. SprintRay is evaluating subscription models to soften the roughly $15,000 upfronthardware cost, an approach explicitly aimed at making the technology accessible to general practices serving middle- and lower-income communities rather than only early-adopter specialty clinics. That’s a meaningful gesture when a company designs financing around volume practices instead of premium cosmetic clinics; it’s betting the technology is ready for the mainstream, not just the showcase.

SprintRay has also been expanding the surrounding ecosystem. Its Midas Digital Press platform uses a capsule-based “Digital Press Stereolithography” approach — resin cartridges that resemble single-use coffee pods — to print crowns in under ten minutes, and the company has previewed multi-unit production capabilities at the 2026 Chicago Midwinter Meeting, with enlarged build capsules intended to let practices fabricate several crowns, inlays, onlays, and veneers in one run rather than one unit at a time. A new “HT” (High Translucency) version of its Ceramic Crown resin specifically targets the chalky, flat appearance that has been a recurring criticism of earlier printed-resin restorations.

Why “Ceramic” Printed Crowns Aren’t Actually Ceramic Crowns

While there are several “permanent crown ceramic” resins out there, here are some details that might get lost in the enthusiasm: The materials industry leaders like SprintRay, Formlabs, Keystone Industries (RODIN), and Ceradirect use for printed crowns are not zirconia or lithium disilicate. They’re resin-matrix composites — biocompatible polymer bases loaded with 50 percent or more ceramic filler, typically silanized dental glass or barium aluminum silicate particles bonded into the resin and cured instantly under 405nm light. This is fundamentally different chemistry from pure ceramics, which rely on a glass matrix with crystalline phases like lithium disilicate or zirconia, an apolycrystalline metal oxide with no glass phase at all — sometimes nicknamed “ceramic steel.”

The practical consequence is strength. These reinforced-resin crowns test at roughly 112to 150 MPa of flexural strength. Lithium disilicate ceramics run 400 to 500 MPa. Zirconia exceeds 1,000 MPa, sometimes reaching 1,200. For context, normal human chewing generates roughly 100 to 150 MPa of force on the back molars, meaning printed resin crowns operate close to the edge of everyday biting forces rather than with the wide safety margin zirconia provides. Clinical forums and dental community feedback reflect this: dentists who’ve tried stretching the material beyond single units — attempting multi-unit bridges, for instance — report high failure and breakage rates, and the material demands a strict minimum wall thickness of 1.0 to 1.5 mm, meaning conservative tooth preparations with thinner margins carry a real fracture risk.

None of this makes the technology useless, but it does make it use-case-specific. The speed and cost advantage is dramatic: a printed resin crown can go from scan to seated restoration in 15 to 45 minutes for a few dollars in material cost, versus a 6-hour-plus mill-and-sinter cycle for zirconia or roughly 45 minutes of milling plus crystallization firing for lithium disilicate. That makes printed resin crowns well suited to single units with moderate bite forces, same-day veneers, and immediate temporaries, while zirconia remains the better choice for heavy grinders, molars under high occlusal load, and bridges, and lithium disilicate still holds the edge for anterior esthetics where light transmission matters most.

Ceramics Are Catching Up on Speed, Too

The clearance race isn’t confined to resin. Researchers at UT Dallas published work in Ceramics International demonstrating same-day chairside 3D-printed zirconia crowns, with debinding — the process of burning out organic binders before sintering — completed in under 30 minutes, rather than the 20-to-100-hour cycles typically required for printed ceramics. If that approach scales beyond the lab, it would narrow the speed gap that has kept true zirconia restorations out of single-visit dentistry, potentially combining zirconia’s mechanical strength with a turnaround time closer to that of resin.

Material jetting is advancing on the prosthetics side as well. 3D Systems secured full EUMDR certification for its NextDent Jetted Dentures workflow, clearing the way for a European launch in summer 2026, and reflecting a broader shift from milling toward additive manufacturing for full-arch removable prosthetics. Meanwhile, Axtra3D and Keystone Industries co-developed KeyModel Ultra Ivory, a non-chipping dental model resin validated specifically for Axtra3D’s Lumia X1 platform — part of a wider industry pattern in which resin and hardware are increasingly co-validated as matched pairs rather than sold as interchangeable, open components, which vendors argue improves clinical predictability but also tends to lock practices into single-vendor ecosystems.

So, Is It Real?

Chairside 3D printing of permanent restorations is real in the sense that FDA-cleared materials and hardware now exist, practices are adopting them, and the economics are increasingly built for general dentistry rather than boutique clinics. It is not yet a wholesale replacement for milled ceramics or sintered zirconia. Clinicians on the front lines remain measured in their enthusiasm. As Nabeel Cajee, DDS, of Advanced Dentistry of Newport Beach, puts it:

“I’m still hesitant to use printed resins for definitive restorations. Zirconia and traditional ceramics remain the gold standard for long-term reliability and aesthetic demand, and I’m not fully convinced the materials are quite there yet. That said, even if the physics never fully match zirconia or glass ceramics, the sheer speed and convenience of chairside fabrication may shift both dentist and patient preferences regardless — patients love same-day results, and that’s a powerful force. This is definitely technology to watch as it matures.”

The honest summary is that dentistry now has a genuine third option alongside milling and lab-based ceramics — fast, cheap, and adequate for the right indication, but not interchangeable with stronger, slower materials for cases that demand them. The technology to watch through the rest of 2026 is whether reinforced resins close the strength gap, whether zirconia closes the speed gap, or whether both simply settle into complementary roles based on where in the mouth — and how fast — a restoration is actually needed.

References:

https://www.voxelmatters.com/sprintray-gets-fda-clearance-to-3d-print-porcelain-dental-crowns

Single-step thermal debinding for ceramics vat photopolymerization in less than 30 minutes

https://www.3dsystems.com/press-releases/3d-systems-secures-class-iia-eu-mdr-certification-nextdentr-jetted-denture-solution

https://3dprintingindustry.com/news/axtra3d-launches-keymodel-ultra-ivory-for-lumia-x1-in-dental-materials-partnership-with-keystone-251592/

The post  Is Chairside 3D-Printed Crown Finally a Reality? appeared first on 3DHeals.

It is already midsummer, but if you missed the latest numbers focusing on early-stage healthcare investing, here are some highlights that could be relevant to entrepreneurs and fundraising startups in the deep tech, medtech, and biopharma space. The first quarter of 2025 marked a pivotal moment for the venture capital and healthcare startup landscape. As uncertainty continued to ripple through global financial markets, investors, founders, and analysts closely monitored funding trends, exit prospects, and innovation hotspots. Drawing on the latest data from leading industry reports—including the HSBC Healthcare Annual Report Q1 2025, PitchBook-NVCA Venture Monitor, Carta’s VC Fund Performance Report, and Wilson Sonsini’s Entrepreneurs Report—we break down the most important takeaways for startups, investors, and ecosystem players. Please find the downloadable versions of these reports at the end of this summary “Reference” section.

⚠️ Disclaimer:

This article is for educational and informational purposes only. The views expressed do not constitute medical or financial advice.

The VC Market: Early Signs of Recovery, but Headwinds Remain

Capital Deployment and Fundraising Focus

2025 began with VC activity rebounding from recent lows, though the environment remains bifurcated:

• Deal Value Rebounds, Exit Options Limited: According to PitchBook-NVCA, Q1 2025 saw $91.5 billion in deal value, reaching highs not seen since early 2022, fueled by mega-rounds in AI, biotech, and infrastructure. Despite this, exit channels remain constrained—with just 12 VC-backed IPOs and most M&A deals concentrating on earlier-stage or distressed assets.
• AI & Biotech Dominate Large Rounds: Over 71% of Q1’s VC dollars flowed into AI and machine learning startups, but healthtech and biopharma had robust showings. Healthcare investment grew 30% year-over-year, with biopharma accounting for nearly half of all healthcare VC dollars (HSBC).
• VC Fund Performance—Selective and Patient: Carta’s report on 2,500 US VC funds reveals the majority of vintages (2019-2024) are still waiting for strong liquidity events, with DPI (distributions) lagging and LPs focusing on managers who’ve proven they can return capital under tough conditions.

Valuations and Down Rounds

• Valuations Hold in Early Stage, Later Stages Correct: Seed and Series A median pre-money valuations (~$17M and $49M, respectively, Wilson Sonsini) remained stable or grew. In contrast, Series C and later-stage valuations contracted, reflecting reality checks after frothy 2021-2022 levels.
• Down Rounds Now the Norm: For post-Series A companies, down rounds made up 29% of deals in Q1—a trend that’s held steady for two years. Liquidation preferences and protective investor terms have become more prevalent (Wilson Sonsini).
• SAFE Financings Now Dominant: Pre-Seed SAFEs appeared in 91% of pre-Seed rounds, with a median raise of $700,000, signaling continued appetite for fast, founder-friendly early-stage mechanisms.

Biopharma & Healthtech: Funding Rebounds Amid Scientific Progress

Growth in Mega Rounds and First Financings

• Mega Rounds Are the Story: Biopharma saw a surge in $100M+ private rounds, often involving large syndicates—including VCs, crossovers, and corporates—reflecting a “safety-in-numbers” approach.
• Focus on Platform, Oncology, Metabolic, and Autoimmune: Platform technologies and oncology drew the biggest checks, followed by metabolic and autoimmune innovations. Notably, the majority of new first-financing deals were preclinical or Phase I stage.

Exits: Still a Long Road

• Tough IPO Market: New IPOs for biopharma in Q1 2025 typically underperformed; post-IPO performance for several 2024-2025 listings (e.g., Metsera, Maze Tx, Sionna Tx, Aardvark Tx) showed share prices down sharply from issue price(>-70%!!), a warning sign for new entrants.
• Private M&A Trends: On the M&A front, only five notable VC-backed biopharma deals occurred in Q1, with two achieving $1B+ up-fronts. The median time to exit via M&A stayed just above four years from first venture round.

VC Operations: What Founders and GPs Should Know

Fundraising and LP Sentiment

• Dry Powder and Fund Size Trends: Carta analysis confirms that fund sizes are skewing smaller, as LPs are cautious amid slow distributions. VC fundraising in the US set a pace for the lowest annual total in a decade.
• LPs Demand Distributions: With just 37% of 2019 vintage funds having made any distributions, managers face intense pressure to deliver DPI before raising new funds.

Deal Terms Rein in Risk

• Investor Protection: Series B and later financings more frequently featured pay-to-play and participating liquidation preferences to protect against portfolio downside.
• Convertible Note and SAFE Trends: Convertible notes are now mostly used as post-seed bridge financing; median raise has increased, and maturity periods have shortened as investors look for earlier liquidity.

Outlook: What to Watch in the Second Half of 2025

• IPO and M&A Windows: Market observers expect windows for IPOs and M&A to re-open modestly as macro policy clarity returns, but the “flight to quality” and the concentration of proceeds in a handful of mega-deals will likely persist.
• AI and Biotech Remain Attractive: Both AI and biotech venture activity are likely to outpace other sectors, particularly for companies with differentiated technology or potential to disrupt major indications.
• Preparation Is Key: With long timeframes to exit, smart startups are preparing for both M&A and IPO dual tracks and keeping close tabs on metrics like cash runway, cap table management, and exit strategy flexibility.

References:

https://carta.com/data/vc-fund-performance-q1-2025

https://www.wsgr.com/a/web/vgnKwoeAkUKT6MTc96jYem/entrepreneurs-report-q1-2025.pdf

https://nvca.org/wp-content/uploads/2025/04/Q1-2025-PitchBook-NVCA-Venture-Monitor-19001.pdf

https://www.foley.com/wp-content/uploads/2025/04/HSBC-2024-HC-Annual-Report-Bio-Foley-Event-4.16.25-002.pdf

The post Navigating 2025: What’s Ahead for Healthcare Venture Capital and Startups appeared first on 3DHeals.

The medical field is undergoing a revolutionary transformation, driven by two cutting-edge technologies: Artificial Intelligence (AI) and 3D printing. When these forces collide, they unleash unparalleled potential for innovation, personalization, and improved patient outcomes. From custom prosthetics to intricate anatomical models for surgical planning, the synergy between AI and 3D printing is reshaping healthcare as we know it. Let’s dive into some of the latest advancements, drawing insights from pioneering researchers and practitioners at the forefront of this exciting intersection.

Personalized Prosthetics for a Better Quality of Life

Creating personalized prosthetics is one of the most compelling applications of AI and 3D printing in healthcare. Merel van der Stelt, a PhD student at 3D Lab Radboudumc, is dedicated to developing prostheses for low- and middle-income countries, utilizing AI for optimized socket shape design (LinkedIn activity). By leveraging AI algorithms, researchers can analyze individual patient data, such as limb shape and movement patterns, to create prosthetics that fit perfectly and function seamlessly. This level of customization enhances comfort, mobility, and overall quality of life for those in need, especially in regions with limited resources.

Moreover, Johann Reinhard, a Research Scientist at Fraunhofer IGD, is pushing the boundaries of 3D printing for eye implants. Using AI to design prints from optical coherence tomography (OCT) images, researchers can create bespoke prosthetic eyes that closely match the patient’s anatomy (New Scientist article). Imagine a future where individuals with eye injuries or congenital disabilities can receive custom-made implants that restore both function and aesthetics. This level of precision and personalization is made possible by AI, which can analyze complex medical images and generate intricate 3D designs.

Enhancing Surgical Planning and Precision

Surgical planning is another area where the combination of AI and 3D printing is making significant strides. Kaiyan Qiu, an assistant professor at Washington State University, uses AI to determine optimal 3D printing parameters for creating surgical planning organ models (WSU press release). Surgeons can generate highly accurate 3D models of patient-specific organs by feeding medical imaging data into AI algorithms. These models allow for detailed pre-operative planning, enabling surgeons to visualize complex anatomical structures and practice procedures before stepping into the operating room. This reduces surgical time and risk and improves patient outcomes by ensuring greater precision and predictability.

Furthermore, Gade Jong, another Assistant Professor, focuses on AI and 3D technologies for anatomical segmentation (Nature article). Precise segmentation of organs and tissues from medical images is crucial for creating accurate 3D models. AI algorithms can automate and refine this process, allowing quicker and more reliable generation of models used in surgical planning. By automating these complex processes, AI enables healthcare professionals to be more efficient and reduce the risk of human error.

Innovations in Bioprinting and Tissue Engineering

Bioprinting, the process of 3D printing living tissues and organs, is a realm where AI is proving invaluable. Ben Kiratitanaporn, a recent PhD graduate, is utilizing AI for 3D cell scaffolds. Cell scaffolds provide the structural support for cells to grow and form functional tissues. AI can optimize the design of these scaffolds, ensuring the proper porosity, mechanical properties, and biocompatibility. Similarly, Guo-Dong Goh, a Research Fellow, focuses on AI for 3D printing tissue-like anatomical models and anomaly detection (ScienceDirect article). AI algorithms can analyze the quality of printed tissues, detecting defects or inconsistencies that might compromise their function.

Associate Professors like Stefanie Mueller at MIT explore human-computer interaction technologies fabricated using 3D printing and AI (MIT news), while Washburn Lab at Carnegie Mellon develops AI for choosing design parameters in the bioprinting of hydrogels (ACS Biomaterials article). Additionally, Demaria at Centro Piaggio, University of Pisa, uses AI to select printing parameters for bioprinting (IJB article). These collective efforts are pushing the boundaries of what’s possible, making the creation of functional tissues and organs closer to reality than ever before.

Drug Delivery and Microneedle Technology

The intersection of AI and 3D printing also opens up exciting possibilities in drug delivery. Moe Elbadawi, a Lecturer, investigates AI, 3D printing, and drug delivery, building upon the work of Alvaro Goyanes at FABRX (ScienceDirect article). By 3D printing personalized drug formulations and delivery devices, we can ensure that patients receive the proper medication in the correct dosage, tailored to their unique needs. This precision approach can improve treatment efficacy and minimize side effects.

Furthermore, Stasoglu at Koç University is leveraging AI to tune 3D printing parameters for microneedle design (MDPI article). Microneedles offer a painless and efficient way to deliver drugs through the skin. AI can optimize the design and fabrication of these tiny needles, enhancing their effectiveness and patient comfort.

Advancements in Wearable Technology and Physiological Monitoring

Wearable technology is another area benefiting from the AI-3D printing synergy. Wei Gao, a professor at Caltech, developed 3D-printed electronic wearable skin that uses AI for physiological monitoring (Science article). These innovative devices can continuously track vital signs and other health indicators, providing valuable data for personalized medicine and early disease detection. Imagine a world where wearable sensors seamlessly integrate with the body, continuously monitoring health and sending alerts when needed.

Defect Detection and Quality Control

Amedeo Bonatti (ORCID) uses AI, particularly expert systems, for defect detection in electron beam melted implants (Liebertpub article). Ensuring the quality and structural integrity of 3D-printed medical devices is critical for patient safety. AI algorithms can analyze 3D-printed objects in detail, identifying any defects or anomalies that might compromise their function. This level of quality control ensures that medical devices meet the highest standards of safety and efficacy.

Devin Roach at Oregon State University also applies AI to 3D printing for biomedical applications. His co-authored article titled “Invertible Neural Networks for Real-Time Control of Extrusion Additive Manufacturing” explores the application of machine learning, specifically invertible neural networks (INNs), to enhance the precision and adaptability of direct ink write (DIW) 3D printing processes. This research is particularly relevant to healthcare due to its potential to improve the fabrication of customized medical devices and implants.(ScienceDirect) Integrating INNs allows for real-time monitoring and optimization of the printing process, ensuring that these medical products meet stringent quality and performance standards. By enabling adaptive control during fabrication, this approach can lead to more reliable and efficient production of complex biomedical structures, ultimately enhancing patient outcomes.

The Future Outlook

The advancements discussed above represent the exciting possibilities when AI and 3D printing intersect in healthcare. As these technologies evolve, we can expect even more groundbreaking innovations. The ability to create personalized medical devices, plan complex surgeries precisely, bioprint functional tissues, and develop advanced drug delivery systems will revolutionize patient care. The integration of AI enhances this process through analysis, automation, and control to ensure that we receive the highest level of patient care. This exciting field has tremendous potential to transform the healthcare landscape and improve lives worldwide.

To learn more about these exciting developments, don’t miss our upcoming virtual event focusing on this very topic.

Keywords: AI in healthcare, 3D printing, personalized medicine, bioprinting, surgical planning, drug delivery, wearable technology.

Related links:

• Artificial Intelligence and Machine Learning in 3D Printing, AR/VR Explores how AI and machine learning are optimizing design, predictive maintenance, quality control, and patient-specific modeling for 3D-printed medical devices and implants. The article also discusses how AI/ML can reduce costs and improve patient outcomes.
• Artificial Intelligence in Healthcare 3D Printing (Webinar) Recap and resources from a theme-based webinar featuring industry and academic leaders discussing the latest AI-driven advancements in healthcare 3D printing.
• When Artificial Intelligence Meets 3D Printing An in-depth article explaining the basics of AI and machine learning, their relevance to 3D printing, and the challenges of integrating AI into real-time 3D printing monitoring and quality control.

Additional 3DHEALS Resources:

1. https://3dheals.com/product/artificial-intelligence-and-machine-learning-in-3d-printing-ar-vr/
2. https://3dheals.com/machine-learning-in-3d-printing-and-bioprinting-a-collection-of-recent-publications/
3. https://3dheals.com/artificial-intelligence-for-segmentation/
4. https://3dheals.com/product/artificial-intelligence-in-healthcare-3d-printing/
5. https://3dheals.com/courses/artificial-intelligence-in-healthcare-3d-printing/
6. https://3dheals.com/artificial-intelligence-updates-for-3d-printing-and-bioprinting/
7. https://3dheals.com/when-artificial-intelligence-meets-3d-printing/
8. https://3dheals.com/tag/machine-learning/
9. https://3dheals.com/artificial-intelligence-and-3d-printing/
10. https://www.youtube.com/c/3DHEALSINNOVATION
11. https://3dheals.com/interview-with-kerim-genc-the-power-of-artificial-intelligence-and-3d-printing/

The post AI in Healthcare 3D Printing: The Future is Now appeared first on 3DHeals.

3D printing and scanning technology has improved since its introduction in 1986, when the first stereolithographic (SLA) systems were introduced [1] and are embraced by both surgical and forensic departments across the world [2]. Despite the applications of 3D printed prosthetics in both medicine and various disciplines of forensic science to date, limited studies can be found on the use of the application in Forensic Medicine, specifically, during post mortem reconstruction. The aim of this article is to explore current reconstruction techniques in forensic medicine to improve aesthetics and bony structure stability in situations requiring repair of skull and facial bone damage due to trauma or the tissue retrieval process in post mortem procedures.

Skull Clips For Repair of Adult Crania After Brain Retrieval

The skull clip drawing (figure 1.1) was created from a 3D model using Onshape (Boston, MA) cloud-based CAD software.

One male adult patient (deceased) and one female adult patient (deceased) between the ages of 60 and 80 were selected for this pilot, which was conducted between April and July 2022. Once the post mortem examination had taken place and the brain retrieval was completed the reconstruction of the skull commenced. Using a 2.5mm hex drill bit, holes in the calvarium were drilled into the bone where the skull clip would be positioned (figure 1.2A). The pre-drilled holes were drilled at the left and right superior temporal region, and posterior occipital region in the area where the clips will be fixed. One side of pre-drilled holes were located on the ‘skull cap’ and the other hole was drilled on the fixed cranium in preparation to fix the skull clip in place. 

While still in a pilot phase, there are positive indications that the new method is faster than the current method of suturing the skull cap back on using muscular anchor points. Importantly, the clips ensure no movement of the skull cap when it is positioned back on the skull, suggesting that the new reconstruction technique is stable and as the skull cap cannot be displaced, ensures no opportunity for disfigurement post reconstruction or prior to family viewing. After completion of the scalp reconstruction took place, an assessment of the forehead region showed no visual resemblance of the skull clips underneath the skin (figure 1.3).

Development and Application of 3D Printed Scaffolds for Paediatric Cranial Post Mortem Reconstruction

The design and prototype phase of the paediatric scaffold began with the use of raw CT imaging data sets from a 27-day old male and 7-month-old female infant. Each DICOM (Digital Imaging and Communications in Medicine) format CT scan, comprising a set of trans axial slices, was imported into the freeware package Meshmixer (Autodesk, Inc, San Francisco, CA) for conversion into a 3D model (OBJ format, Wavefront Technologies, Santa Barbara, CA).  The resulting 3D skull model was then imported into the freeware Computer Aided Design (CAD) package Onshape (Boston, MA) where it was used in the development of a parametric ‘Boolean’ cranial cavity model (figure 2.1).  

The ‘Boolean’ Digital Model

Based on the Boolean operation of union, which takes 2 objects and merges the interiors of the object creating a new object or shape, [3] it was realised that the cranial cavity could be suitably represented by a series of five spheres located in the sagittal plane to infill the internal space (figure 2.2).

The diameters and sagittal plane locations of the spheres could be easily adjusted to fill the internal space in CAD depending upon the dimensions of the OBJ (Wavefront OBJect) skull model.

Once the scaffold model was finalised in CAD it was then exported in Standard Tessellation Language (STL) format for conversion into a 3D-printable file.  

3D Printed Skull Replica

Prior to printing the Boolean scaffold a replica skull was required to test the prototype. The 7-month-old female data set was selected for the first replica skull model as access to the cranial cavity was required in order to place the scaffold inside the cavity. 

The 7-month-old female skull replica was printed in PLA plastic (figure 2.3). The scaffold would then be placed inside this replica and the skull cap would be placed back on top of the scaffold to assess size and suitability of infilling the cranial cavity.

The Boolean System Scaffold

Two types of plastic were selected for use: i) Poly Lactic Acid (PLA) or “cornstarch” and ii) Acrylonitrile Butadiene Styrene (ABS). The stability and flexibility/rigidity of each plastic type were compared in each prototype. The print time for the Boolean scaffold was 8 hours for each prototype. 

Results – Boolean System

Prototype 1

Prototype 1 was printed in PLA plastic using the following parameters: 210C extrusion temperature, 60C build plate temperature and 0.25mm layer thickness. 

The resulting 3D printed model (figure 2.4) was inflexible and the fit inside the cranial cavity was too tight. These factors would not allow for the calvarium to be replaced neatly, with the scaffold pushing the calvarium away from the cranial base. This would likely create deformities, as the 2 separate cranial pieces could not be united during the reconstruction process.

Prototype 2

The goal for prototype 2 was to create a scaffold that had more flexibility than the first prototype. It was decided to proceed with ABS plastic for all remaining Boolean prototypes, with the hope that more flexible plastic will facilitate better reconstruction once the calvarium was replaced. 

This second prototype was printed with the following parameters: 230C extrusion temperature, 110C build plate temperature and 0.25mm layer thickness. These printing parameters would remain consistent for all subsequent ABS prototypes.

The resulting scaffolding was still too rigid to permit a seamless union of the 2 cranial pieces during reconstruction. This would likely pose a problem with anatomical asymmetry in an actual postmortem case, as the pressure of the scaffold inside the cranial cavity on the infant cranial bones could cause the cranial bones to deform. 

Prototypes 3 – 5

Significant changes were made to prototypes 3-5. Rather than having a solid structure (Boolean shaped) that formed the scaffold, various engineered sections were cut and removed in CAD to increase the compliance and flexibility of the mould in the cranial cavity and in turn, create a better fit to infill the cavity, resulting in a seamless union of the calvarium to the cranial base. 

Prototype 5 was the final Boolean inspired scaffold produced. The engineered ‘fronds’ from the previous prototype 3 and 4 were cut closer together to reduce the over-expansion of the scaffold (see figure 2.7 a) and 2.7 b)). 

Prototype 5, however was still too inflexible and tight for the printed skull replica. Its rigidity continued to cause the printed skull replica to push outwards and the calvarium could not be placed with anatomical symmetry.

Band System

The band system was a more generic approach and much easier to create than the Boolean system, the band system incorporates a 3D printed band with a self-locking sawtooth rachet-type mechanism (figures 2.8a, 2.8b, 2.8c). The 27-day old male skull replica was used for testing the band scaffold. The print time for each single band was 4 hours.

Prototype 1

Prototype 1 of the band system comprised of a single band scaffold conforming to the coronal plane of the internal cranial cavity 3D CAD model (figure 2.9a and 2.9b). 

This protype was printed using PLA. Although the geometry of the prototype 1 band scaffold was similar to the shape of the internal skull cavity, in practice the PLA material proved to be too difficult to manipulate when being placed in the skull replica and adjusting the size the of the band and the saw-toothed mechanised failed to lock in to one another. This resulted in the band pushing against the petals of the infants’ skull from the inside, resulting in large gaps between the individual cranial bones (figure 2.9b) and would likely cause deformities in skull shape during reconstruction. 

Prototype 2

Minor adjustments were made for prototype 2 to enhance the built-in saw-toothed ratchet mechanisms and improve flexibility of the band. To improve flexibility, PLA was replaced with ABS plastic.

Changes to the ratchet mechanism were made, making them bigger and longer to enable a larger range of diameter to be accommodated and the self-locking mechanism to be more functional (figure 2.10). This prototype generally worked well in the coronal plane, with no observable pressure on the cranial petals, but the sagittal plane skull components were not supported.  In practice, this could lead to collapse of deformity of the cranial bones in the midline plane of the skull. It was evident that a band system in multiple anatomical planes would be needed to improve the stability of the petals from inside the cranial cavity.

Prototype 3

To support the sagittal skull components an additional band was created to enable a more complete scaffold system for the cranial cavity. It also involved the printing of an additional base plate with a stabilising clip that the sagittal band could slide through creating the coronal and sagittal scaffold band in one piece (figure 2.11).

Prototype 3 was again printed in ABS plastic for flexibility, had a total print time of 8 hours which included both bands and the base plate and clip. This prototype incorporated ratchet-band system in both the sagittal and coronal planes, enabling a more complete support of the cranial bones from within. The multiaxial bands were key to providing support without creating unnecessary deformations of the fragile infant cranial bones and this design became the final prototype for this project. 

Discussion

The use of 3D printed skulls clips would improve the efficiency of the reconstruction method and stability of the calvarium for adult post mortem reconstruction. This is a low cost, time efficient novel technique that has strong potential to become standard practice nationally in the field of post mortem cranial reconstruction.

Due to the fragility of infant cranial bones, this research sought to design and implement a stable scaffold that would improve cranial reconstruction outcomes in paediatric postmortems.

Whilst the Boolean system proved to be time consuming to print and difficult to fit the band system however, specifically the double band which provided structural integrity in both the coronal and sagittal planes, showed more promise, as the ratchet system design means that the band size can be modified to suit the size of the infants’ cranial cavity at the time of reconstruction. 

There are no current guidelines as to what materials can be used during reconstruction. It is therefore recommended that the band system be considered as a new method of infant reconstruction and use of this reconstruction method should be included in the training of future forensic postmortem technicians. 

The results of this study demonstrate that 3D printing has potential to improve post mortem reconstruction outcomes in the discipline of Forensic Medicine and should be considered as a standard approach to cranial reconstruction, particularly in infants. 

About the Author:

Lisa Bilton is a Forensic Medicine Researcher with a drive to enhance and optimise professional standards in post mortem reconstruction using innovative technology such as 3D printing, supporting partnerships across medico-legal teams and driving interdisciplinary collaboration with forensic scientists/pathologists and physicians.

Having completed more than 500 post mortem reconstructions as a Forensic Scientist/Technician within Australian Health Pathology, Lisa is exploring opportunities to extend her Forensics research in 3D printing in post mortem reconstruction. 

She firmly believes that with the technology available today, can improve the standards in industry and – most importantly – provide better outcomes for families of deceased, particularly paediatric patients.

Alongside her forensic medicine practice and research, Lisa has also lectured mortuary practise subjects with Western Sydney University.

References:

Aimar, A., Palermo, A., & Innocenti, B. (2019). The role of 3D printing in medical applications: A State of the Art. Journal of Healthcare Engineering. 1:5340616.

Carew, R., Morgan, R., Phil, D., & Rando, C. (2019). A Preliminary Investigation into the accuracy of 3D modelling and 3D printing in forensic anthropology evidence reconstruction.  Journal of Forensic Sciences, (64), 342–352.

Charton, J., Laurentjoye,M., & Youngjun, K. (2017). 3D Boolean operations in virtual surgical planning. International Journal of CARS, (12),1697–1709

The post 3D Printing in Post Mortem Reconstruction appeared first on 3DHeals.