Methodological challenges in machine learning and deep learning applied to food analysis: A critical review
- 1Department of Drug Science and Technology, University of Turin, Via Giuria 9, 10124 Turin, Italy
- 2Department of Drug Science and Technology, University of Turin, Via Giuria 9, 10124 Turin, Italy. Electronic address: [email protected]
Abstract
Artificial Intelligence (AI) is increasingly applied for food quality control, authenticity assessment, and chemical profiling. However, the reliability and industrial applicability of Machine Learning (ML) and Deep Learning (DL) models critically depend on how the datasets are constructed, validated, and interpreted. Among the different analytical techniques used in food analysis, chromatographic fingerprints and chromatographic hyphenated techniques typically generate high-dimensional datasets characterised by structured peak distributions, retention time variability, and strong multicollinearity among adjacent variables, making model development and validation more challenging. These features, combined with limited sample sizes typical of food studies, pose significant challenges for model generalisation and transferability across analytical batches and industrial settings. This work provides a critical evaluation of methodological practices in ML and DL, supported by a systematic review of 165 peer-reviewed studies published between 2015 and 2025. Particular attention is given to common issues such as class imbalance, the misuse of analytical replicates, and the widespread reliance on internal cross-validation without independent test sets. To experimentally assess these factors, a chromatographic dataset on coffee origins was used as a case study to examine how unbalanced class distributions, varying training set sizes, and replicate handling affect model performance. Our results show that high-fold cross-validation and Leave-One-Out Cross-Validation consistently overestimate model accuracy compared to external validation, while treating analytical replicates as independent samples artificially inflates performance without improving true predictive power. Learning curves reveal that most ML models plateau with moderate training sizes, whereas DL models require substantially larger datasets to avoid overfitting. The findings highlight that rigorous validation strategies and careful management of chromatographic data are essential for developing robust, reproducible, and industrially transferable ML models in food chromatography.
Related
- Predicting CYP450 Induction and PXR Activation with Machine Learning Approaches: Challenges and Opportunities
- Harnessing Machine Learning for Accelerated Drug Discovery: Opportunities and Unmet Challenges
- Machine Learning Identifies an Information-Sufficiency Regime for Critical-Mineral Recovery
- From prediction to process control: a critical review of machine learning in reactive dyeing of cotton
- Docking, Molecular Dynamics, QSAR, and Machine Learning in Alzheimer's Small-Molecule Drug Discovery: Current Utility, Key Limitations, and Reproducibility Challenges
BibTeX
@article{felizzato2026methodological,
title = {{Methodological challenges in machine learning and deep learning applied to food analysis: A critical review}},
author = {Giorgio Felizzato and Eloisa Bagnulo and Giulia Tapparo and Giorgia Botta and Chiara Emilia Irma Cordero and Erica Liberto and Andrea Caratti},
journal = {Journal of Chromatography A},
year = {2026},
month = jun,
doi = {10.1016/j.chroma.2026.467176},
url = {https://doi.org/10.1016/j.chroma.2026.467176},
note = {CC BY},
}
RIS
TY - JOUR
TI - Methodological challenges in machine learning and deep learning applied to food analysis: A critical review
AU - Giorgio Felizzato
AU - Eloisa Bagnulo
AU - Giulia Tapparo
AU - Giorgia Botta
AU - Chiara Emilia Irma Cordero
AU - Erica Liberto
AU - Andrea Caratti
PY - 2026
DA - 2026/06/10
JO - Journal of Chromatography A
DO - 10.1016/j.chroma.2026.467176
UR - https://doi.org/10.1016/j.chroma.2026.467176
AB - Artificial Intelligence (AI) is increasingly applied for food quality control, authenticity assessment, and chemical profiling. However, the reliability and industrial applicability of Machine Learning (ML) and Deep Learning (DL) models critically depend on how the datasets are constructed, validated, and interpreted. Among the different analytical techniques used in food analysis, chromatographic fingerprints and chromatographic hyphenated techniques typically generate high-dimensional datasets characterised by structured peak distributions, retention time variability, and strong multicollinearity among adjacent variables, making model development and validation more challenging. These features, combined with limited sample sizes typical of food studies, pose significant challenges for model generalisation and transferability across analytical batches and industrial settings. This work provides a critical evaluation of methodological practices in ML and DL, supported by a systematic review of 165 peer-reviewed studies published between 2015 and 2025. Particular attention is given to common issues such as class imbalance, the misuse of analytical replicates, and the widespread reliance on internal cross-validation without independent test sets. To experimentally assess these factors, a chromatographic dataset on coffee origins was used as a case study to examine how unbalanced class distributions, varying training set sizes, and replicate handling affect model performance. Our results show that high-fold cross-validation and Leave-One-Out Cross-Validation consistently overestimate model accuracy compared to external validation, while treating analytical replicates as independent samples artificially inflates performance without improving true predictive power. Learning curves reveal that most ML models plateau with moderate training sizes, whereas DL models require substantially larger datasets to avoid overfitting. The findings highlight that rigorous validation strategies and careful management of chromatographic data are essential for developing robust, reproducible, and industrially transferable ML models in food chromatography.
SN - 0021-9673
SN - 1873-3778
ER -