Scholarlink

<p>Food adulteration and contamination remain critical global challenges for supply-chain integrity, public health protection, and consumer trust. Conventional destructive analytical techniques such as high-performance liquid chromatography (HPLC) and gas chromatography–mass spectrometry (GC–MS), while accurate, are invasive, time-consuming, and poorly suited for high-throughput or in-line monitoring. This review evaluates terahertz (THz) spectroscopy and imaging, in combination with artificial intelligence (AI), as emerging non-destructive testing (NDT) modalities for agri-food quality and safety, with emphasis on their integration into processing and quality-control systems. A systematic synthesis of studies published between 2021 and 2025 examines THz applications in adulterant detection, spoilage monitoring, and quality assessment, benchmarked against established techniques including hyperspectral imaging, Raman, near- and mid-infrared spectroscopy, ultrasound, and X-ray inspection.​ Recent reports demonstrate classification accuracies exceeding 90% for selected dry-food adulteration tasks, regression errors below 0.05&#xa0;g/100&#xa0;g for quantitative analyses, and signal-to-noise improvements of approximately 25–40% using AI-assisted THz processing pipelines. Persistent challenges including limited penetration in high-moisture foods, dataset scarcity, calibration transfer, and model generalisation are critically discussed. Building on these insights, the review outlines industrial and regulatory pathways aligned with HACCP-based systems and ISO/AOAC frameworks, and discusses socio-economic implications such as food-waste reduction, resource efficiency, and equitable compliance verification across global agri-food supply chains.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Next-generation food safety: Terahertz and AI for non-destructive testing of food products

  • 期刊论文

  • 发表日期：

    2026年3月8日

查看全文
  • E. Vinodhini Shanmugapriya,
  • A. Shirly Edward,
  • A. Mercy Latha

摘要

Food adulteration and contamination remain critical global challenges for supply-chain integrity, public health protection, and consumer trust. Conventional destructive analytical techniques such as high-performance liquid chromatography (HPLC) and gas chromatography–mass spectrometry (GC–MS), while accurate, are invasive, time-consuming, and poorly suited for high-throughput or in-line monitoring. This review evaluates terahertz (THz) spectroscopy and imaging, in combination with artificial intelligence (AI), as emerging non-destructive testing (NDT) modalities for agri-food quality and safety, with emphasis on their integration into processing and quality-control systems. A systematic synthesis of studies published between 2021 and 2025 examines THz applications in adulterant detection, spoilage monitoring, and quality assessment, benchmarked against established techniques including hyperspectral imaging, Raman, near- and mid-infrared spectroscopy, ultrasound, and X-ray inspection.​ Recent reports demonstrate classification accuracies exceeding 90% for selected dry-food adulteration tasks, regression errors below 0.05 g/100 g for quantitative analyses, and signal-to-noise improvements of approximately 25–40% using AI-assisted THz processing pipelines. Persistent challenges including limited penetration in high-moisture foods, dataset scarcity, calibration transfer, and model generalisation are critically discussed. Building on these insights, the review outlines industrial and regulatory pathways aligned with HACCP-based systems and ISO/AOAC frameworks, and discusses socio-economic implications such as food-waste reduction, resource efficiency, and equitable compliance verification across global agri-food supply chains.

Graphical Abstract