Microplastic Contamination and Detection in Food Systems: A Review of Machine Learning, Traditional Methods, and Other Relevant Factors

Microplastics (MPs) are widespread contaminants in food, beverages, and drinking water, raising concerns over potential health risks. Accurate and standardized detection a significant analytical chemistry challenge due to complex food matrices and limitations in traditional methods like Fourier-Transform Infrared (FT-IR) and Raman spectroscopy, scanning electron microscopy (SEM), and pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS), etc. These techniques often require intensive sample preparation and struggle to detect small or low-abundance particles. Recently, machine learning (ML) has emerged as a promising solution to automate detection, improve accuracy, and handle complex datasets. This review outlines the major contamination pathways of MPs, evaluates current detection technologies, and emphasizes the role of ML-particularly deep learning, hyperspectral imaging, and AIenhanced spectroscopy-as transformative tools in microplastic analysis. Key challenges such as data scarcity, matrix interference, and lack of standardization are discussed, along with opportunities for real-time detection, robotics integration, and open-access datasets. By combining analytical chemistry with AI, this review highlights the potential of ML-based approaches to enhance food safety and support policy development for microplastic contamination.