Is my food safe? - AI-based classification of lentil flour samples with trace levels of gluten or nuts.

Food Chem

Departamento de Ingeniería Química y de Materiales, Universidad Complutense de Madrid, 28040 Madrid, Spain. Electronic address:

Published: August 2022

An artificial intelligence-based method to rapidly detect adulterated lentil flour in real time is presented. Mathematical models based on convolutional neural networks and transfer learning (viz., ResNet34) have been trained to identify lentil flour samples that contain trace levels of wheat (gluten) or pistachios (nuts), aiding two relevant populations (people with celiac disease and with nut allergies, respectively). The technique is based on the analysis of photographs taken by a simple reflex camera and further classification into groups assigned to adulterant type and amount (up to 50 ppm). Two different algorithms were trained, one per adulterant, using a total of 2200 images for each neural network. Using blind sets of data (10% of the collected images; initially and randomly separated) to evaluate the performance of the models led to strong performances, as 99.1% of lentil flour samples containing ground pistachio were correctly classified, while 96.4% accuracy was reached to classify the samples containing wheat flour.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.foodchem.2022.132832DOI Listing

Publication Analysis

Top Keywords

lentil flour
16
flour samples
12
samples trace
8
trace levels
8
flour
5
food safe?
4
safe? ai-based
4
ai-based classification
4
lentil
4
classification lentil
4

Similar Publications

Starch-rich faba bean, yellow lentil, and yellow field pea flours were subjected to submerged fermentation using Aspergillus oryzae and Lactobacillus plantarum starter mono- or co-cultures, to increase protein contents of the flours. Fermentation mixes were supplemented with up to 35 g/L urea, ammonium sulfate and/or monoammonium phosphate as nitrogen sources. Protein contents of the flours increased 2-2.

View Article and Find Full Text PDF

Characterization and In Vitro Digestion Kinetics of Purified Pulse Starches: Implications on Bread Formulation.

Foods

January 2025

Whistler Center for Carbohydrate Research and Department of Food Science, Purdue University, West Lafayette, IN 47907, USA.

This study investigated the contribution of pulse starches (PSs) to the slowly digestible starch (SDS) properties observed in pulses. Purified pulse starches from 17 commonly consumed pulses were examined, focusing on their digestion kinetics using a pancreatic alpha-amylase (PAA) and rat intestinal acetone powder (RIAP) mixture. Chickpea starch, exhibiting a slow digestibility profile, was incorporated as an ingredient to confer slow digestibility to refined wheat flour bread.

View Article and Find Full Text PDF

This paper aimed to study the nutritional, phytochemical and rheological properties of some composite flours based on wheat flour (WF) mixed with non-germinated (LF) and sprouted lentil flour (SLF), in order to fortify the wheat flour and to obtain functional bakery/pastry products. The composite flours based on wheat flour and bean lentil flour (BLWF) and sprouted lentil flour (SLWF) were analyzed from the point of view of proximate composition (proteins, lipids, total carbohydrates, and minerals), content of individual and total polyphenols (TPC), as well as the contents of macro and microelements. For use in baking/pastries, the composite flours were tested from the point of view of rheological behavior using the MIXOLAB system, and the profiles obtained were compared with those of bread and biscuit.

View Article and Find Full Text PDF

Periweissella beninensis LMG 25373, belonging to the recently established Periweissella genus, exhibits unique motility and high adhesion capabilities, indicating significant probiotic potential, including resilience under simulated gastrointestinal conditions. This study demonstrates for the first time that P. beninensis LMG 25373^T produces a dextran-type exopolysaccharide (EPS) with a distinctive high degree of branching (approximately 71 % of α-(1 → 6)-linkages and 29 % α-(1 → 3)-linkages).

View Article and Find Full Text PDF

In view of the increasing demand for plant-based protein alternatives, along with the increased protein requirements of older adults, the formulation of new food concepts based on protein-rich ingredients from legumes or pseudo-cereals seems a promising approach. Previous studies have found that solid-state fermentation can improve the nutritional value and digestibility of plant-based commodities; however, scarce evidence exists regarding the effect on prebiotic potential. This study aimed to compare the effect of fermented and unfermented quinoa and lentil flours on the colonic microbiota, as well as that of new food prototypes (gels and breads) made with the flours.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!