On-site Food Authenticity Testing: Advances in Miniaturization of Spectrometers and Machine Learning

Abstract

Laboratory-based quality checks, although accurate, are often time consuming and expensive and can only be performed by trained scientific personnel, which is a major drawback in a fast-moving supply chain of perishable food commodities. Vibrational spectroscopy is a proven technique for the non-destructive and rapid measurement of food quality and authenticity. Table-top vibrational spectroscopy equipment such as Fourier transform infrared, near-infrared and Raman spectrometers are already used widely for food authentication and quality analysis. However, recent advances in miniaturized sensors supported with artificial intelligence (AI) and the internet of things (IoT) might take the technique out of the laboratory and into the hands of food processors, farmers and consumers. In this chapter, the principle of food authenticity testing using miniature spectrometers is described and several chemometric data analysis tools, software and machine learning algorithms that can potentially be used or are being used for food authenticity testing using miniature vibrational spectroscopy devices are discussed. A summary of commercially available portable miniature spectrometers for food testing is given and major start-up initiatives that use miniature vibrational spectroscopy sensors and AI for food analysis are presented. Finally, an overview of the application of various miniature and hand-held spectroscopy devices and sensors for food authenticity testing is given.