Semi-Autonomic AI LF-NMR Sensor for Industrial Prediction of Edible Oil Oxidation Status

Abstract

The evaluation of an oil’s oxidation status during industrial production is highly important for monitoring the oil’s purity and nutritional value during production, transportation, storage, and cooking. The oil and food industry is seeking a real-time, non-destructive, rapid, robust, and low-cost sensor for nutritional oil’s material characterization. Towards this goal, a 1H LF-NMR relaxation sensor application based on the chemical and structural profiling of non-oxidized and oxidized oils was developed. This study dealt with a relatively large-scale oil oxidation database, which included crude data of a 1H LF-NMR relaxation curve, and its reconstruction into T1 and T2 spectral fingerprints, self-diffusion coefficient D, and conventional standard chemical test results. This study used a convolutional neural network (CNN) that was trained to classify T2 relaxation curves into three ordinal classes representing three different oil oxidation levels (non-oxidized, partial oxidation, and high level of oxidation). Supervised learning was used on the T2 signals paired with the ground-truth labels of oxidation values as per conventional chemical lab oxidation tests. The test data results (not used for training) show a high classification accuracy (95%). The proposed AI method integrates a large training set, an LF-NMR sensor, and a machine learning program that meets the requirements of the oil and food industry and can be further developed for other applications.

Novel Contactless Sensors for Food, Beverage and Packaging Evaluation

“…This special issue (SI) focused on the novel methods developed using contactless sensors for food, beverages and packaging. This SI is composed of nine papers related to the use of spectral analysis using different methods, such as near-infrared spectroscopy to classify eggs into cage or free-range production practices [1], a semi-automatic low field nuclear magnetic resonance (LF-NMR) coupled with machine learning modelling to predict oxidation in edible oil [2], the use of vibrational spectroscopy to measure ethanol and methanol levels in pisco [3], and the use of an infrared laser sensor to monitor the gasphase CO 2 in Champagne headspace when swirling [4]. Other papers focus on the use of other specific sensors, such as a low-cost and portable electronic nose (e-nose) to predict aromas and roasting intensity in coffee [5] and the use of flexible sensors for monitoring oyster survival rates [6].…”

“…The authors successfully classified samples in cage and free-range production with an overall accuracy of 76% for whole eggs and 86% for egg whites and yolks separately. Osheter et al [2] reported high accuracy (95%) on the classification of oxidation levels in edible oil using LF-NMR as inputs of convolutional neural networks (CNN) supervised machine learning modelling. In another publication, Menevseoglu et al [3] used a Raman and Fourier-transform infrared spectroscopy handheld device to assess ethanol and methanol in Pisco through the bottle.…”