Answers are grounded in this paper and cite their sources.
Evaluates machine learning-enhanced FTIR for detecting gluten cross-contamination in flours, suggesting practical applications for food safety.
Gluten-related disorders can result in serious health issues if not managed properly by maintaining a 100% gluten-free diet. In this study, FTIR coupled with supervised machine-learning (ML) approaches was evaluated for the rapid detection and quantification of cross-contamination between a non-gluten (corn flour (CF)) and gluten-rich (wheat flour (WF), barley flour (BF) and rye flour (RF)) flours, at contamination levels of 0–10% (w/w), with 0.5% increments. Attenuated total reflectance (ATR) spectra were acquired and subjected to multiple preprocessing treatments before ML model development. The F1-scores (0.963, 0.949, 0.963 and 1.0), R2p (0.96, 0.94, and 0.98), and RMSEP (0.82, 0.99, and 0.53) obtained for the best results show that the methods used have the potential to authenticate the cross-contact of non-gluten and gluten-rich flours within the defined contamination levels. These findings further demonstrate that spectral preprocessing plays a critical role in optimizing model performance, with the Savitzky–Golay derivative and robust auto-scaling among the most effective methods evaluated. This represents an important step in AI-based rapid detection and quantification of gluten contamination in foods, suggesting that, with further development and validation, this approach may eventually be adaptable for integration into portable and handheld FTIR systems to support on-site quality assurance in the grain and food processing industries.
No takes yet. Share an insight, caveat, or question.
Okeke et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: