Scientists have created a rapid system to generate highly detailed 3D models of ants using X-ray imaging, robotics and AI, dramatically accelerating insect morphology research.
 |
| Researchers from University of Maryland and Karlsruhe Institute of Technology have used AI, robotics and synchrotron imaging to rapidly create detailed 3D models of hundreds of ant species. Image: SD/CH |
Karlsruhe, Germany — March 11, 2026
Scientists have developed a high-speed technique to generate extremely detailed three-dimensional models of ants, combining advanced X-ray imaging, robotics and artificial intelligence to accelerate biodiversity research. The breakthrough enables researchers to reconstruct hundreds of insect species in digital form far faster than previously possible.
The study, published on March 5 in the journal Nature Methods, was led by researchers including Evan Economo of the University of Maryland and Thomas van de Kamp of the Karlsruhe Institute of Technology in Karlsruhe, Germany.
For years, scientists studying insect morphology—the analysis of physical structures—have relied on micro-CT scanners to produce detailed internal images of specimens. While the technology provides microscopic resolution, scanning a single insect can take up to ten hours, limiting the scale of research that can be conducted.
To overcome this bottleneck, the research team developed a high-throughput system that integrates a synchrotron particle accelerator with automated robotics and AI-based image processing. The project, known as Antscan, allows scientists to rapidly scan thousands of ant specimens and convert the data into interactive 3D digital models.
Using facilities at the Karlsruhe Institute of Technology, researchers scanned around 2,000 preserved ants in just one week. Completing a similar dataset using conventional CT scanning methods would have required roughly six years of continuous work.
The specimens—preserved in ethanol and collected from museums, research institutions and private specialists worldwide—were transported to Germany for imaging. A powerful synchrotron beam generated intense X-rays capable of penetrating multiple samples quickly, producing detailed cross-sectional images.
Automation played a crucial role in accelerating the process. A robotic system handled the specimens during scanning, rotating each sample and replacing it with another every 30 seconds. The scanning process produced stacks of two-dimensional X-ray images that were later combined into high-resolution 3D reconstructions.
However, many of the ants initially appeared distorted because of how they were positioned during scanning. To address this, computer science students developed AI tools capable of automatically correcting the posture of the insects, generating realistic digital models that resemble ants in natural positions.
The resulting models reveal internal anatomical features—including muscles, nervous systems, digestive organs and stingers—at micrometer-level resolution. Researchers say the digital reconstructions can also be animated or placed into virtual reality environments, expanding their potential use in scientific research, education and digital media.
The Antscan database has already supported additional scientific discoveries. In a separate study published in December 2025 in Science Advances, researchers used the dataset to analyze how ant colonies balance worker size and physical strength.
By examining more than 500 species, scientists found that colonies investing less in thick exoskeleton armour often maintain larger worker populations. The findings suggest that reducing investment in protective cuticle may allow colonies to expand more effectively and diversify their workforce.
The ability to measure structural features such as cuticle volume with high precision has been particularly valuable for evolutionary research, as such measurements were previously difficult to calculate accurately.
Researchers believe the growing Antscan archive could eventually function as a comprehensive digital library of ant biodiversity. Beyond taxonomy and evolutionary studies, the scans could also be used to train machine-learning systems capable of automatically identifying ant species during field research.
Looking ahead, the team plans to expand the database by scanning additional specimens and applying the same combination of robotics, high-energy imaging and artificial intelligence to other biological datasets. Scientists say the approach could open new avenues for exploring the diversity of life on Earth while preserving digital records of species for future generations.