Full text loading...
Antibiotic residues in food products and environmental matrices pose significant public health risks, including antimicrobial resistance and toxicological effects. Traditional detection methods face limitations regarding sensitivity, cost-effectiveness, and field applicability, necessitating advanced technological solutions. A systematic literature review was conducted, examining publications from 2020 to 2024 using PubMed and academic databases. Keywords included “Artificial Intelligence,” “Machine Learning,” “Antibiotic Residue Detection,” “Biosensors,” “Spectroscopy,” and “Food Safety.” Studies integrating AI/ML with biosensors, optical systems, and electrochemical platforms were analysed. AI-enhanced detection systems demonstrated superior performance metrics. Electrochemical sensors with gradient boosting algorithms achieved a 99% classification accuracy for antibiotic identification. Machine learning-powered optical immunosensors achieved detection limits of 0.03-0.4 ng/mL for the simultaneous quantification of multiple antibiotics. Convolutional Neural Networks resolved spectral overlaps with R2 values exceeding 0.984, while smartphone-based systems enabled portable detection with high precision and recall metrics. AI/ML integration significantly improves sensitivity, specificity, and multiplexing capabilities over conventional methods. These technologies enable real-time, on-site monitoring and address spectral interference challenges. However, standardisation protocols and cross-matrix validation remain critical gaps, requiring further research. AI/ML technologies represent a paradigm shift in antibiotic residue analysis, offering enhanced detection capabilities for food safety and environmental monitoring. Continued development of robust, standardised AI models is essential for regulatory adoption and widespread implementation in public health protection.
Article metrics loading...
Full text loading...
References
Data & Media loading...