Unlocking biodiversity data with robotic imaging and AI-driven transcription of natural history collections
Science Advances · 01.05.2026 · 12 (18)
DOI: 10.1126/sciadv.aeb6975 ↗Kurz gefasst
Fachartikel in der Zeitschrift Science Advances, erschienen am 01.05.2026. Verfasst von Christine A Johnson und 19 weiteren Autor:innen. Beteiligte Einrichtung: Division of Invertebrate Zoology, American Museum of Natural History, New York, NY, USA. Hauptthemen laut MEDLINE: Biodiversity, Robotics, Artificial Intelligence. Die Arbeit nennt 9 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: National Science Foundation Advancing Digitization of Biological Collections Marine Invertebrate Thematic Collections Network, National Science Foundation Advancing Digitization of Biological Collections Thematic Collections Network iDigBees, National Science Foundation DBI Partner to an Existing Network Grant.
Abstract (Originalsprache)
Digitizing metadata on natural history specimen labels remains a critical bottleneck for biodiversity research. We present a transformative workflow integrating robotic imaging with artificial intelligence (AI)–driven transcription for rapid, comprehensive data extraction from specimen labels. Single high-resolution images of specimens and associated labels were submitted to Gemini 2.5 Flash and GPT-4 Turbo to extract verbatim textual information. This approach yielded ~600 verbatim transcriptions per hour, a 30-fold increase in efficiency compared to traditional manual methods, which yielded ~20 transcriptions per hour. Releasing historical specimen metadata facilitates information accessibility and provides temporal and spatial context for a variety of analyses. Our method fosters the reconnection of disparate biological datasets previously segregated among departments or institutions to unite ecologically interdependent components (e.g., host/parasite and pollinator/plant) for a more complete understanding of biodiversity dynamics.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Division of Invertebrate Zoology, American Museum of Natural History, New York, NY, USA.
- Columbia University, New York, NY, USA.
ORCID-Profile: Christine A Johnson, Erin M Willigan, Parker M Austin, Kyle L Chang, Jovanni M Gonzalez, Isobel E J Mifsud, Kitri A Miller, Marinela Nicolescu, Ana Maria Ruiz, Elliott Y Ruo, Natalia C Zurita, Lily F Berniker, Estefanía Rodríguez
Förderung
- National Science Foundation Advancing Digitization of Biological Collections Marine Invertebrate Thematic Collections Network
- National Science Foundation Advancing Digitization of Biological Collections Thematic Collections Network iDigBees
- National Science Foundation DBI Partner to an Existing Network Grant
Zitieren
APA
Johnson, C. A., Willigan, E. M., Hwang, E. Y., Austin, P. M., Chang, K. L., Daly, P., Davis, A., Gonzalez, J. M., Kao, C. N., Mifsud, I. E. J., Miller, K. A., Nicolescu, M., Ruiz, A. M., Ruo, E. Y., Schwartzman, J. L., Seeman, A. B., Stenzel, J., Zurita, N. C., Berniker, L. F., & Rodríguez, E. (2026). Unlocking biodiversity data with robotic imaging and AI-driven transcription of natural history collections. Science Advances, 12(18). https://doi.org/10.1126/sciadv.aeb6975
Vancouver
Johnson CA, Willigan EM, Hwang EY, Austin PM, Chang KL, Daly P, et al. Unlocking biodiversity data with robotic imaging and AI-driven transcription of natural history collections. Science Advances. 2026;12(18). doi:10.1126/sciadv.aeb6975
BibTeX
@article{Johnson2026,
author = {Johnson, Christine A and Willigan, Erin M and Hwang, Ellen Y and Austin, Parker M and Chang, Kyle L and Daly, Peter and Davis, Ann and Gonzalez, Jovanni M and Kao, Chia-Ni and Mifsud, Isobel E J and Miller, Kitri A and Nicolescu, Marinela and Ruiz, Ana Maria and Ruo, Elliott Y and Schwartzman, Julia L and Seeman, Alyssa B and Stenzel, Janice and Zurita, Natalia C and Berniker, Lily F and Rodríguez, Estefanía},
title = {Unlocking biodiversity data with robotic imaging and AI-driven transcription of natural history collections},
journal = {Science Advances},
year = {2026},
volume = {12},
number = {18},
doi = {10.1126/sciadv.aeb6975}
}
EUVIMED