EUVIMEDOpdag EUVIMED
Crossref

AI for difficult herbarium specimens: identification of peat mosses (subgenus Sphagnum ) without dissection

Damon P Little iD, Blanka Aguero iD, A Jonathan Shaw iD, Michael Tessler iD

New Phytologist · 2026-07 · 251 (2) · 782-799

DOI: 10.1111/nph.70461 ↗

At a glance

Journal article in New Phytologist, published online on 2025-08-16. Written by Damon P Little and 3 co-authors. Institution: Center for Biodiversity & Evolution, New York Botanical Garden, Bronx, NY, 10458, USA. Major MEDLINE topics: Sphagnopsida, Artificial Intelligence. Cited 5 times so far (Crossref); 118 references. Funded by: Directorate for Biological Sciences.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Artificial intelligence (AI) for image‐based herbarium specimen identification has thus far focused on plants that can be identified by eye. Here, we develop the first AI focused on identifying herbarium specimens of a bryophyte group, peat mosses in Sphagnum subgenus Sphagnum . These plants have substantial morphological plasticity, and confident identifications require time‐consuming dissections and microscopy. We hypothesized that AI, using unmagnified low‐resolution images, can (H 1 ) identify species and (H 2 ) discover novel morphological characters. We collected 4386 publicly available herbarium specimen images of all 10 North American species and imaged an additional 105 specimens with determinations verified by DNA and morphology. AI identification was generally successful with our newly formulated FireNetSEz model (68% AUCPR (area under the curve: precision recall)). We produced a reduced dataset (the five most imaged species) that we, the authors, could attempt. Our identifications took hours and were all lower‐scoring than the AI. These H 1 results show that AI can learn hard‐to‐identify botanical species without microscopy and outperform both generalist botanists and Sphagnum experts. Regarding H 2 , we found the AI focuses on edges of organs that humans often ignore. AI holds promise for hard botanical identifications and the potential to rapidly identify Sphagnum , which is important for studying peatlands that strongly impact climate.

Subjects

MeSH terms (MEDLINE)

SphagnopsidaArtificial IntelligenceSpecies Specificity

Author keywords

Sphagnumartificial intelligencebotanical specimensherbariummoss

Authors and institutions

  • Center for Biodiversity & Evolution, New York Botanical Garden, Bronx, NY, 10458, USA.
  • Graduate Center, City University of New York, New York, NY, 10016, USA.
  • Department of Biology & L. E. Anderson Bryophyte Herbarium, Duke University, Durham, NC, 27708, USA.
  • Department of Biology, Medgar Evers College, City University of New York, Brooklyn, NY, 11225, USA.
  • Institute for Comparative Genomics, American Museum of Natural History, New York, NY, 10024, USA.

ORCID profiles: Damon P Little, Blanka Aguero, A Jonathan Shaw, Michael Tessler

Funding

  • Directorate for Biological Sciences

Cite

APA

Little, D. P., Aguero, B., Shaw, A. J., & Tessler, M. (2026). AI for difficult herbarium specimens: identification of peat mosses (subgenus Sphagnum ) without dissection. New Phytologist, 251(2), 782-799. https://doi.org/10.1111/nph.70461

Vancouver

Little DP, Aguero B, Shaw AJ, Tessler M. AI for difficult herbarium specimens: identification of peat mosses (subgenus Sphagnum ) without dissection. New Phytologist. 2026;251(2):782-799. doi:10.1111/nph.70461

BibTeX
@article{Little2026,
  author = {Little, Damon P and Aguero, Blanka and Shaw, A Jonathan and Tessler, Michael},
  title = {AI for difficult herbarium specimens: identification of peat mosses (subgenus Sphagnum ) without dissection},
  journal = {New Phytologist},
  year = {2026},
  volume = {251},
  number = {2},
  pages = {782--799},
  doi = {10.1111/nph.70461}
}
Start for free →