Researchers have used a synchrotron particle accelerator, robotics, and AI to create high-resolution 3D models of ants from 800 species. The project scanned 2,000 specimens in just one week, far faster than traditional methods. This effort, called Antscan, aims to build a digital library of insect biodiversity.
For over a decade, Evan Economo's lab at the University of Maryland has used micro CT scanners to image insects, focusing on their morphology. However, these scans are slow, often taking 10 hours per specimen, as Economo noted: "One limitation is that you can get this rich 3D dataset, but it could take 10 hours to scan one specimen."
In a study published on March 5, 2026, in Nature Methods, Economo and Thomas van de Kamp from the Karlsruhe Institute of Technology (KIT) in Germany led a team that accelerated the process. By combining a synchrotron accelerator's intense X-ray beam with robotics and AI, they scanned 2,000 ethanol-preserved ant specimens from global museums in one week. A robotic changer rotated each specimen every 30 seconds, generating stacks of 2D images converted into 3D models.
Julian Katzke, the study's first author and a former student at the Okinawa Institute of Science and Technology (OIST), explained the efficiency: "We've estimated that if we were to carry out this project with a lab-based CT scanner, it would take six years of continuous operation. With the setup at KIT, we scanned 2,000 specimens in a single week."
Initial scans showed ants in awkward positions, so University of Maryland computer science students in James Purtilo's course developed AI for pose estimation to make models appear natural. Purtilo described it as: "A capstone is intended to challenge students to integrate skills, function as an effective team and demonstrate their ability to solve real problems. And this problem was a doozy."
The Antscan models reveal microscopic details like muscles, nervous systems, and stingers at micrometer resolution. Raw data is publicly available, with an online viewer for exploration. Economo emphasized broader impact: "The value of this study is not only about ants -- it's much broader. When specimens are digitized, we can build libraries of organisms that can streamline their use from scientific laboratories to classrooms to Hollywood studios."
Antscan data supported a December 19, 2025, Science Advances paper by Economo and others, analyzing over 500 ant species. It found a negative correlation between exoskeleton cuticle volume and colony size, linking physical traits to evolutionary success. This builds on a June 2025 Cell study co-authored by Economo on ant genomes.
Economo plans to expand the database, stating: "This work moves us further into the big data era of capturing, analyzing and sharing organismal shape and form." The paper, titled "High-throughput phenomics of global ant biodiversity," highlights potential for AI in biodiversity research.
