QDs fluorescent immunosensor based on magnetic separation coupled with machine learning for aflatoxin B1 detection in...

Area of Science:

• Analytical Chemistry
• Biotechnology
• Food Safety

Background:

• Aflatoxin B1 (AFB1) is a prevalent mycotoxin in vegetable oils, posing significant public health risks.
• Current detection methods often lack the required speed and sensitivity for effective monitoring.

Purpose of the Study:

• To develop a novel competitive immunofluorescent biosensor for rapid and sensitive AFB1 detection.
• To evaluate the biosensor's performance and applicability in real-world samples, such as peanut oil.

Main Methods:

• Synthesis and characterization of antibody-conjugated nanoparticles and QDs-AFB1-BSA probes using FT-IR spectroscopy and zeta-potential measurements.
• Optimization of a competitive immunofluorescent assay for AFB1 detection.
• Application of machine learning (RF_FE method) for spectral data analysis.

Main Results:

• The developed biosensor achieved a low detection limit (0.0995 ngmL^-1) and a broad linear range (0.01–100 ngmL^-1).
• The assay demonstrated high selectivity, reproducibility, and excellent recovery rates in spiked peanut oil samples.
• Machine learning analysis showed superior predictive capabilities with low errors, correlating spectral data with AFB1 concentrations.

Conclusions:

• The novel competitive immunofluorescent biosensor is a promising tool for sensitive and rapid AFB1 detection in vegetable oils.
• The integration of machine learning enhances the accuracy and reliability of the analytical method.
• This approach contributes to improved food safety and public health surveillance for mycotoxin contamination.