Predicting Mycotoxin Contamination in Irish Oats Using Deep and Transfer Learning

Mycotoxin contamination poses a significant risk to cereal crop quality, food safety, and agricultural productivity. Accurate prediction of mycotoxin levels can support early intervention strategies and reduce economic losses. This study investigates the use of neural networks and transfer learning models to predict mycotoxin contamination in Irish oat crops as a multi-response prediction task. Our dataset comprises oat samples collected in Ireland, containing a mix of environmental, agronomic, and geographical predictors. Five modelling approaches were evaluated: a baseline multilayer perceptron (MLP), an MLP with pre-training, and three transfer learning models; TabPFN, TabNet, and FT-Transformer. Model performance was evaluated using regression (RMSE, R 2 ) and classification (AUC, F1) metrics, with results reported per toxin and on average. Additionally, permutation-based variable importance analysis was conducted to identify the most influential predictors across both prediction tasks. The transfer learning approach TabPFN provided the overall best performance (RMSE = 0.65, R 2 = 0.86, F1 = 0.92, AUC = 0.98), followed closely by the baseline MLP. Our variable importance analysis revealed that weather history patterns in the 90-day pre-harvest period were the most important predictors, alongside grain moisture content.

Authors

Institutions

Publication Details

Journal
npj Science of Food
Published
2026-11-19
DOI
https://doi.org/10.1038/s41538-026-01208-5
Citations
3
Primary Topic
Mycotoxins in Agriculture and Food
Type
article
Field-Weighted Citation Impact
6.82

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Predicting Mycotoxin Contamination in Irish Oats Using Deep and Transfer Learning

Adam Gauley, Anne P. Nugent, Steven Kildea, Brett Greer et al.
3 citations
npj Science of Food
Mycotoxins in Agriculture and Food
6.82
article

Predicting Mycotoxin Contamination in Irish Oats Using Deep and Transfer Learning

Adam Gauley, Anne P. Nugent, Steven Kildea, Brett Greer, Martin Danaher, Lisa Black, Breige A. McNulty, Melissa Di Rocco, Fiona M. Doohan, Diana Bucur, Andrew C. Parnell, Subramani Natarajan, Alan Inglis, Julie P. Meneely, Chris Elliott, Naoise McKenna
article en
3 citations

Abstract

Mycotoxin contamination poses a significant risk to cereal crop quality, food safety, and agricultural productivity. Accurate prediction of mycotoxin levels can support early intervention strategies and reduce economic losses. This study investigates the use of neural networks and transfer learning models to predict mycotoxin contamination in Irish oat crops as a multi-response prediction task. Our dataset comprises oat samples collected in Ireland, containing a mix of environmental, agronomic, and geographical predictors. Five modelling approaches were evaluated: a baseline multilayer perceptron (MLP), an MLP with pre-training, and three transfer learning models; TabPFN, TabNet, and FT-Transformer. Model performance was evaluated using regression (RMSE, R 2 ) and classification (AUC, F1) metrics, with results reported per toxin and on average. Additionally, permutation-based variable importance analysis was conducted to identify the most influential predictors across both prediction tasks. The transfer learning approach TabPFN provided the overall best performance (RMSE = 0.65, R 2 = 0.86, F1 = 0.92, AUC = 0.98), followed closely by the baseline MLP. Our variable importance analysis revealed that weather history patterns in the 90-day pre-harvest period were the most important predictors, alongside grain moisture content.

npj Science of Food
University College Dublin (IE), Queen's University Belfast (GB), Teagasc - The Irish Agriculture and Food Development Authority (IE), National University of Ireland, Maynooth (IE), Food for Health Ireland (IE), Agri Food and Biosciences Institute (GB)
Department of Agriculture, Environment and Rural Affairs, UK Government, Department of Agriculture, Food and the Marine, Ireland
Zero hunger
Openalex Percentile: Top 5%
Mycotoxins in Agriculture and Food
6.82
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon