Automated identification of fossil benthic  foraminifera from the Peruvian margin using convolutional neural networks

Hayat, Sikandar; Mojtahid, Meryem; Elliot, Mary; Cardich, Jorge; Geslin, Emmanuelle; de Garidel-Thoron, Thibault; Carré, Matthieu; Barras, Christine

doi:10.5194/jm-45-95-2026

Articles | Volume 45, issue 1

https://doi.org/10.5194/jm-45-95-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

Advances and challenges in modern and benthic foraminifera...

https://doi.org/10.5194/jm-45-95-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 45, issue 1

Research article

|

26 Jan 2026

Research article |

| 26 Jan 2026

Automated identification of fossil benthic foraminifera from the Peruvian margin using convolutional neural networks

Sikandar Hayat, Meryem Mojtahid, Mary Elliot, Jorge Cardich, Emmanuelle Geslin, Thibault de Garidel-Thoron, Matthieu Carré, and Christine Barras

Data sets

Automated and Manual counts of fossil benthic foraminifera for 31 sedimentary samples from Peruvian margin sedimentary cores Sikandar Hayat et al. https://doi.org/10.17882/108821

Benthic foraminifera images to train the CNN model (Automated identification of fossil benthic foraminifera from the Peruvian margin using convolutional neural networks) Sikandar Hayat et al. https://doi.org/10.5281/zenodo.17965134

Short summary

We studied how artificial intelligence can speed up the study of tiny ocean organisms called benthic foraminifera, whose shells help scientists understand past ocean changes. Normally, identifying and counting them under a microscope takes a lot of time and skill. Here, we compared automated image analysis and human counting for 31 samples. The automated method showed promising results for faster and reliable research.