Parameter Optimization of Support Vector Machine to Improve the Predictive Performance for Determination of Aflatoxin B1 in Peanuts by Olfactory Visualization Technique

Array-optimized artificial olfactory sensor enabling cost-effective and non-destructive detection of mycotoxin-contaminated maize

The MobileNetV3-based improved sine-cosine algorithm (ISCA-MobileNetV3) was combined with an artificial olfactory sensor (AOS) to address the redundancy in olfactory arrays, thereby achieving low-cost and high-precision detection of mycotoxin-contaminated maize. Specifically, volatile organic compounds of maize interacted with unoptimized AOS containing eight porphyrins and eight dye-attached nanocomposites to obtain the scent fingerprints for constructing the initial data set. The optimal decision model was MobileNetV3, with more than 98.5% classification accuracy, and its output training loss would be input into the optimizer ISCA. Remarkably, the number of olfactory arrays was reduced from 16 to 6 by ISCA-MobileNetV3 with about a 1% decrease in classification accuracy. Additionally, the developed system showed that each online evaluation was less than one second on average, demonstrating outstanding real-time performance for ensuring food safety. Therefore, AOS combined with ISCA-MobileNetV3 will encourage the development of an affordable and on-site platform for maize quality detection.

Determination of procymidone residues in rapeseed oil based on olfactory visualization technology

A new means about olfactory visualization technique for the quantitative analysis of procymidone residues in rapeseed oil has been proposed. First, an olfactory visualization system was set up to collect volatile odor information from rapeseed oil samples containing different concentrations of procymidone residues. Then, we utilized four intelligent optimization algorithms, namely particle swarm optimization (PSO), genetic algorithm (GA), ant colony optimization (ACO) and simulated annealing (SA), to optimize the characteristics of the sensors. Finally, support vector machine regression (SVR) models employing optimized features were constructed for the quantitative detection of procymidone residues in rapeseed oil. The study demonstrated that the SA-SVR model demonstrated superior prediction results, achieving a high determination coefficient of prediction (R P 2 ) at 0.9894. As indicated by the results, it is possible to successfully conduct non-destructive detection of procymidone residues in edible oil by the olfactory visualization technology.

Detection of the Inoculated Fermentation Process of Apo Pickle Based on a Colorimetric Sensor Array Method

Apo pickle is a traditional Chinese fermented vegetable. However, the traditional fermentation process of Apo pickle is slow, easy to ruin, and cannot be judged with regard to time. To improve fermentation, LP-165 (L. Plantarum), which has a high salt tolerance, acidification, and growth capacity, was chosen as the starter culture. Meanwhile, a colorimetric sensor array (CSA) sensitive to pickle volatile compounds was developed to differentiate Apo pickles at varying degrees of fermentation. The color components were extracted from each dye in the color change profiles and were analyzed using principal component analysis (PCA) and linear discriminant analysis (LDA). The fermentation process of the Apo pickle was classified into four phases by LDA. The accuracy of backward substitution verification was 99% and the accuracy of cross validation was 92.7%. Furthermore, the partial least squares regression (PLSR) showed that data from the CSA were correlated with pH total acid, lactic acid, and volatile acids of the Apo pickle. These results illustrate that the CSA reacts quickly to inoculated Apo pickle and could be used to detect fermentation.