Distinct thermal patterns to detect and quantify trace levels of wheat flour mixed into ground chickpeas. (1st August 2022)

• Record Type: Journal Article
• Authors: Cancilla, John C.; Pradana-López, Sandra; Pérez-Calabuig, Ana M.; López-Ortega, Sandra; Rodrigo, Carlos; Torrecilla, José S.
• Is Part Of: Food chemistry. Volume 384(2022)
• Publication Date: 2022-08-01
• Subjects: Gluten -- Food fraud -- Thermography -- Transfer learning -- Residual neural network; Food -- Analysis -- Periodicals; Food -- Composition -- Periodicals; 664
• DOI: 10.1016/j.foodchem.2022.132468
• Languages: English
• ISSNs: 0308-8146

Abstract

Highlights: Thermographic evolution to quantify trace levels of wheat in chickpea flour. ResNet34 serves as quick quality and health control method for food. Fourteen groups of chickpea flour with 0 to 50 ppm of wheat flour classified. Blinded images classified at a 99.1% accuracy displaying quantification ability. Indirect assessment of gluten content as safety measure. Abstract: This paper combines intelligent algorithms based on a residual neural network (ResNet34) to process thermographic images. This integration is aimed at detecting traces of wheat flour, in concentrations from 1 to 50 ppm, mixed into chickpea flour. Using an image database of over 16 thousand samples to train the ResNet34, and 1712 images to blindly test it, the optimized intelligent algorithm is able to classify the thermographic images into 14 classes according to the concentration of wheat flour at a 99.0% correct classification rate. These results open the door to the development of a simple, fast, and inexpensive prototype that can be used during the entire distribution chain to help protect brands and consumers. The detection and quantification of trace amounts of wheat flour, or indirectly gluten, serves as a quality control and health safety application protecting, for example, people with celiac disease.