.png){kind=icon}
.png){kind=icon}
Return
Characterising Nanoparticulate Food Colourant Dispersions via Laser Scattering Imaging and Machine Learning to Predict Lethality in Caenorhabditis elegans
.png){kind=icon}
PRE
AI
S
S
A
J
R
DOI：10.1007/s11947-026-04525-5.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Nanotechnology in the food industry requires rigorous toxicological safety assessments and precise nanomaterial characterisation. This study evaluates the capacity of laser scattering imaging (LSI) combined with machine learning to predict the in vivo toxicity (lethality) of two nanoparticulated food colourants (Fe₂O₃ and Ag) in Caenorhabditis elegans. Forty-eight dispersions were analysed, considering the three factors: nanoparticle material, concentration, and thermal treatment time. Physicochemical characterisation and lethality assays were integrated with LSI data to train support vector machine (SVM) models. Results showed that thermal treatments significantly altered the particles’ properties and biological impact. The LSI-based SVM models predicted lethality with R2 values > 0.80. Notably, implementing a preliminary classification step to isolate non-effective cases (samples with no significant difference from the control in lethality terms) enhanced predictive performance to R2 > 0.90. This hierarchical approach demonstrates that LSI can serve as a rapid, low-cost, and non-destructive new approach methodology (NAM) for screening the safety of food-grade nanomaterials, potentially reducing the need for extensive in vivo testing.
Keywords:
Food colouring
Nanoparticles
Toxicity
C. elegans
Imaging analysis
Machine learning
Journal
IF:
5.8
Papers:
4.5K
Citations:
1.5W
Organization
Cited Papers