Estimating weaning duration from incremental dentine δ15N and δ13C using a sequence-based LSTM neural network: A deep learning framework for bioarchaeological applications
PLOS One · 2025
DOI: 10.1371/journal.pone.0337619 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Elissavet Ganiatsou und 2 weiteren Autor:innen. Beteiligte Einrichtung: Laboratory of Biological Anthropology, Department of Humanities, Democritus University of Thrace, Komotini, Greece. Hauptthemen laut MEDLINE: Weaning, Carbon Isotopes, Dentin, Nitrogen Isotopes. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Research Council.
Themen: Beatmung und Weaning
Abstract (Originalsprache)
The estimation of weaning duration from incremental dentine δ 15 N and δ 13 C values offers insights into health, nutrition, and demography in past populations. In this study, we developed a novel machine learning approach to estimate weaning duration using a sequence-based Long Short-Term Memory (LSTM) neural network model. The model was trained on published serial isotopic data from 279 individuals across three tooth types (M1, dM1, dM2). We enhanced predictive performance by incorporating temporal features (differences, rolling averages, slopes) and encoding tooth type to allow the model to learn type-specific patterns. The model was evaluated with RMSE, MAE, and R 2 metrics using Monte Carlo dropout to quantify predictive uncertainty. As a proof of concept, we compared our model to established methods in weaning estimation (WEAN, ChangeR) using 20 individuals from the site of Thessaloniki. Our model achieved an RMSE of 0.46 years (95% CI: 0.38–0.54), a MAE of 0.36 years (95% CI: 0.29–0.44), and an R 2 of 0.82 (95% CI: 0.73–0.88), indicating reliable predictive accuracy. In contrast to previous approaches that rely primarily on δ 15 N, our method leverages sequential patterns in both δ 15 N and δ 13 C to infer the weaning process. This study introduces the first machine learning framework specifically designed to estimate weaning duration from serial isotopic data. By modelling complex, non-linear developmental trajectories, it offers a scalable and robust tool for reconstructing weaning practices in bioarchaeological and related research studies.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Laboratory of Biological Anthropology, Department of Humanities, Democritus University of Thrace, Komotini, Greece.
- Institute of Applied Bioarchaeological Research, Democritus University of Thrace, Komotini, Greece.
ORCID-Profile: Elissavet Ganiatsou
Förderung
- European Research Council
Zitieren
APA
Ganiatsou, E., Souleles, A., & Papageorgopoulou, C. (2025). Estimating weaning duration from incremental dentine δ15N and δ13C using a sequence-based LSTM neural network: A deep learning framework for bioarchaeological applications. PLOS One. https://doi.org/10.1371/journal.pone.0337619
Vancouver
Ganiatsou E, Souleles A, Papageorgopoulou C. Estimating weaning duration from incremental dentine δ15N and δ13C using a sequence-based LSTM neural network: A deep learning framework for bioarchaeological applications. PLOS One. 2025. doi:10.1371/journal.pone.0337619
BibTeX
@article{Ganiatsou2025,
author = {Ganiatsou, Elissavet and Souleles, Angelos and Papageorgopoulou, Christina},
title = {Estimating weaning duration from incremental dentine δ15N and δ13C using a sequence-based LSTM neural network: A deep learning framework for bioarchaeological applications},
journal = {PLOS One},
year = {2025},
doi = {10.1371/journal.pone.0337619}
}
EUVIMED