Magnetic Fe

Area of Science:

• Analytical Chemistry
• Food Safety
• Materials Science

Background:

• Quinolone antibiotics are widely used in the dairy industry.
• Excessive antibiotic residues in dairy products pose a significant food safety concern.
• Sensitive and accurate detection methods are crucial for monitoring these contaminants.

Purpose of the Study:

• To develop a novel method for the classification and quantification of three quinolone antibiotics (Ciprofloxacin, Norfloxacin, Levofloxacin) in dairy products.
• To utilize Surface-Enhanced Raman Scattering (SERS) combined with machine learning for enhanced detection.
• To address the challenge of detecting structurally similar antibiotic compounds.

Main Methods:

• Fabrication of a magnetic Covalent Organic Framework (COF)-based SERS substrate.
• Application of SERS for spectral acquisition of quinolone antibiotics.
• Integration of machine learning algorithms (PCA-k-NN, PCA-SVM, PCA-Decision Tree) for data analysis and classification.
• Determination of Limit of Detection (LOD) for individual antibiotics.

Main Results:

• Achieved 100% classification accuracy for the spectral dataset of the three antibiotics.
• Established low Limits of Detection (LOD): Ciprofloxacin (CIP) at 5.61 × 10⁻⁹ M, Levofloxacin (LEV) at 1.44 × 10⁻⁸ M, and Norfloxacin (NFX) at 1.56 × 10⁻⁸ M.
• Demonstrated the effectiveness of the combined SERS and machine learning approach for complex sample analysis.

Conclusions:

• The developed magnetic COF-based SERS substrate coupled with machine learning offers a highly sensitive and accurate method for detecting quinolone antibiotics in dairy products.
• This approach provides a promising new tool for ensuring food safety and combating antibiotic overuse in the dairy industry.
• The method's ability to differentiate between structurally similar molecules highlights its potential for complex residue analysis.