Evan Economo’s lab has been using micro-CT scanners to image insect specimens for more than a decade. These X-ray scans allow scientists to examine the physical structure and form of insects, a field of study known as morphology. Although this technology provides highly detailed 3D data, it is expensive and time-consuming.
“One limitation is that we can obtain this rich 3D dataset, but it can take 10 hours to scan one specimen,” explained Economo, chair of the Department of Entomology at the University of Maryland and the James B. Gahan and Margaret H. Gahan Professor.
In a study published in a journal nature method On March 5, 2026, Economo and colleagues tested a new approach designed to dramatically speed up the process. The project brought together researchers led by Economo and Thomas van de Kamp from Germany’s Karlsruhe Institute of Technology (KIT). Their team combined synchrotron particle accelerators, X-ray imaging, robotics, and artificial intelligence (AI) to produce interactive digital reconstructions representing 800 species of ants.
By combining these technologies, scientists can now scan specimens more quickly and convert raw image data into detailed 3D models.
“We estimated that performing this project using a laboratory-based CT scanner would require six years of continuous operation,” said Julian Katzke, lead author of the study and an alumnus of the Economo lab at Okinawa Institute of Science and Technology Graduate University (OIST). “With the KIT setup, we scanned 2,000 specimens in one week.”
The effort, known as Antscan, could lead to future large-scale digitization projects for many types of living things. The raw data used to build the model is publicly available for download, and users can explore the completed 3D ant online using an integrated viewer.
“The value of this research is not just about ants; it’s much broader,” Professor Economo said. He is currently an adjunct professor at OIST in addition to his UMD role. “Once specimens are digitized, we can build libraries that can streamline the use of living things from science labs to classrooms to Hollywood studios.”
Building a digital library of ant biodiversity
To assemble this extensive digital archive, the research team collected ethanol-preserved ant specimens from museums, partner institutions, and experts around the world. After organizing the samples by species and class, the specimens were transported to KIT for high-throughput micro-CT imaging. This method works similarly to a medical CT scan, but with much higher magnification.
At this facility, a synchrotron particle accelerator produced a powerful X-ray beam that could quickly scan many samples. A robotic sample changer handled the insects during the process, rotating each specimen and replacing it with the next specimen every 30 seconds. This rapid workflow produced a stack of 2D images that the researchers later combined to create a complete 3D model.
Initially, scans caught ants in twisted or awkward positions. These distorted poses were far from the life-like models scientists wanted to create. To address this issue, students in UMD Computer Science Associate Professor James Purtilo’s CMSC 435: “Software Engineering” course began developing an AI tool to automate “pose estimation.” The technology adjusts the scanned images so that the ants appear in their natural position, similar to how they would appear in the wild.
“This collaboration was a great opportunity for us,” said Pultillo. “Capstone is meant to challenge students to demonstrate their ability to integrate skills, work as an effective team, and solve real-world problems. And this problem was outrageous.”
The resulting Antscan model reveals internal details such as muscles, nervous systems, digestive organs, and needles at micrometer-level resolution. These digital ants can also be animated and placed in virtual reality environments for scientific research, education, and entertainment.
“It would take years to do this manually, so without these computational tools we basically would never have been able to do it,” Economo said. “We are now making great strides toward creating a living library of interactive models that address the planet’s biodiversity. AI will enable us to explore the diversity of life and share it with the world.”
Antscan data fuels new research
The ever-growing Antscan database has already proven useful for scientific research. Economo also served as senior author on a paper published in the journal scientific progress In this study, scientists used Antscan data to investigate whether ant colonies benefit from having more small workers or fewer individuals with stronger bodies.
The researchers investigated the relationship between cuticle volume, colony size, and evolutionary diversification in more than 500 species of ants. The cuticle forms the protective outer layer of the ant’s exoskeleton. Its production requires nitrogen and other minerals, so thicker armor means a greater resource investment for the individual ant.
Their analysis revealed a strong negative correlation between cuticle volume and colony size. In other words, colonies that invest less in thicker armor may be able to support more workers, potentially allowing them to grow larger and diversify better.
Antscan made these measurements possible because the 3D model allows researchers to accurately calculate the volume of the cuticle, which was previously difficult to measure. The project also scanned the same ant species investigated in a June 2025 study published in the journal . cell Co-authored with Economo, which has generated a set of high-quality ant genomes. By combining these datasets, scientists can better understand the relationship between physical characteristics and genetic variation.
The scans are so detailed that they could also be useful in training machine learning systems to recognize ants in the field during behavioral studies. Economo plans to continue expanding the database by scanning additional specimens and collaborating with UMD computer science students to apply these AI techniques to new biological datasets.
“This research moves us further into the big data era of capturing, analyzing and sharing the shapes and forms of living things,” Economo said. “The potential for integrating these data with other data types and technologies is enormous and very exciting.”
Their paper “High-throughput phenomics of global ant biodiversity” was published in the journal. nature method March 5, 2026.
This article is an edited version of the text provided by Okinawa Institute of Science and Technology.
This research was supported by the German Ministry of Research and Education. Ministry of Science, Research and the Arts of Baden-Württemberg. German Research Foundation (grant numbers INST 35/1503-1 FUGG and 502787686). Okinawa Institute of Science and Technology Graduate University, Japan Society for the Promotion of Science (grant numbers 18K14768, 21K06326, 22KJ3077) and Australian Research Council (award number IC 180100008). HUN-REN Hungarian Research and National Research, Development and Innovation Fund (grant number K 147781). Conselho Nacional de Desenvolvimento Científico e Tecnológico (grant number 301495/2019-0); a joint initiative of the Critical Ecosystem Partnership Fund, the French Development Agency, Conservation International, the European Union, the Global Environment Facility, the Government of Japan, and the World Bank. National Science Foundation (grant numbers DEB-1932467, DEB 1927161, and IOS-2128304). Italian Ministry of University Research. Hong Kong Environmental Conservation Fund (Award No. Nb. ECF 137/2020). And Fundação para a Ciência ea Tecnologia. This article does not necessarily reflect the views of these organizations.
