Multispectral Imaging Flow Cytometry for Spatio‐Temporal Pollen Trait Variation Measurements of Insect‐Pollinated Plants DOI Creative Commons
F. Walther, Martin Hofmann, Demetra Rákosy

и другие.

Cytometry Part A, Год журнала: 2025, Номер unknown

Опубликована: Апрель 8, 2025

ABSTRACT Artificial intelligence (AI) surpasses human accuracy in identifying ordinary objects, but it is still challenging for AI to be competitive pollen grain identification. One reason this gap the extensive trait variation grains. In classical textbooks, size relies on only 25–50 grains, mostly one plant and site. Lack of databases can cause limited application machine learning approaches real‐world samples. Therefore, our study aims investigate sources spatial temporal morphology fluorescence. For purpose, 64,001 grains from four herbaceous insect‐pollinated species Achillea millefolium L., Lamium album Lathyrus vernus (L.) Bernh., Lotus corniculatus L. sampled across years seven locations Central Germany were measured using multispectral imaging flow cytometry. Observed variations very species‐specific; however, most species, significant differences as well found at least trait. We could also show that variability identity a particular sample influence classifications multiple measurements different origins provide robust AI‐based identifications.

Язык: Английский

Multispectral Imaging Flow Cytometry for Spatio‐Temporal Pollen Trait Variation Measurements of Insect‐Pollinated Plants DOI Creative Commons
F. Walther, Martin Hofmann, Demetra Rákosy

и другие.

Cytometry Part A, Год журнала: 2025, Номер unknown

Опубликована: Апрель 8, 2025

ABSTRACT Artificial intelligence (AI) surpasses human accuracy in identifying ordinary objects, but it is still challenging for AI to be competitive pollen grain identification. One reason this gap the extensive trait variation grains. In classical textbooks, size relies on only 25–50 grains, mostly one plant and site. Lack of databases can cause limited application machine learning approaches real‐world samples. Therefore, our study aims investigate sources spatial temporal morphology fluorescence. For purpose, 64,001 grains from four herbaceous insect‐pollinated species Achillea millefolium L., Lamium album Lathyrus vernus (L.) Bernh., Lotus corniculatus L. sampled across years seven locations Central Germany were measured using multispectral imaging flow cytometry. Observed variations very species‐specific; however, most species, significant differences as well found at least trait. We could also show that variability identity a particular sample influence classifications multiple measurements different origins provide robust AI‐based identifications.

Язык: Английский

Процитировано

0