FT-NIR and linear discriminant analysis to classify chickpea seeds produced with harvest aid chemicals

Unraveling almonds deterioration using whole-cell biosensor coupled with machine learning approaches and SHAP interpretation

As almonds are prone to oxidation during storage, it is essential to construct a real-time method to monitor the quality of almonds efficiently. In this study, the in situ detection was developed using whole-cell biosensor combined with machine learning algorithms. Mantel test between volatile compounds and promoters was conducted to provide theoretical support for luminescence response of whole-cell biosensor. SHAP algorithm was implemented to visualize machine learning models for global and local explanations. As a result, six biosensors of pspA, uvrA, katG, ropS, grpE, and leuA were explored to fabricate whole-cell biosensor. The LDA, LR, and PLS-DA exhibited relatively lower prediction accuracy, while SVM, and RF outperformed the above linear models with the accuracy of 97.5 % and 100 %. Moreover, the whole-cell biosensor array combined with RF algorithm offers a favorable strategy for almond deterioration. This study provides an in situ, efficient, environment-friendly approach for quality assurance in almonds and other food products.

Predicting Perennial Ryegrass Cultivars and the Presence of an Epichloë Endophyte in Seeds Using Near-Infrared Spectroscopy (NIRS)

Perennial ryegrass is an important temperate grass used for forage and turf worldwide. It forms symbiotic relationships with endophytic fungi (endophytes), conferring pasture persistence and resistance to herbivory. Endophyte performance can be influenced by the host genotype, as well as environmental factors such as seed storage conditions. It is therefore critical to confirm seed quality and purity before a seed is sown. DNA-based methods are often used for quality control purposes. Recently, near-infrared spectroscopy (NIRS) coupled with hyperspectral imaging was used to discriminate perennial ryegrass cultivars and endophyte presence in individual seeds. Here, a NIRS-based analysis of bulk seeds was used to develop models for discriminating perennial ryegrass cultivars (Alto, Maxsyn, Trojan and Bronsyn), each hosting a suite of eight to eleven different endophyte strains. Sub-sampling, six per bag of seed, was employed to minimize misclassification error. Using a nested PLS-DA approach, cultivars were classified with an overall accuracy of 94.1-98.6% of sub-samples, whilst endophyte presence or absence was discriminated with overall accuracies between 77.8% and 96.3% of sub-samples. Hierarchical classification models were developed to discriminate bulked seed samples quickly and easily with minimal misclassifications of cultivars (<8.9% of sub-samples) or endophyte status within each cultivar (<11.3% of sub-samples). In all cases, greater than four of the six sub-samples were correctly classified, indicating that innate variation within a bag of seeds can be overcome using this strategy. These models could benefit turf- and pasture-based industries by providing a tool that is easy, cost effective, and can quickly discriminate seed bulks based on cultivar and endophyte content.

Artificial intelligence-driven tool for spectral analysis: identifying pesticide contamination in bees from reflectance profiling

Pesticide poisoning constantly threatens bees as they forage for resources in pesticide-treated crops. This poisoning requires thorough investigation to identify its causes, underscoring the importance of reliable pesticide detection methods for bee monitoring. Infrared spectroscopy provides reflectance data across hundreds of spectral bands (hyperspectral reflectance), presumably enabling the efficient classification of pesticide contamination in bee carcasses using artificial intelligence (AI) models, such as machine learning. In this study, bee contamination by commercial formulations of three insecticides-dimethoate (organophosphate), fipronil (phenylpyrazole), and imidacloprid (neonicotinoid)-as well as glyphosate, the most widely used herbicide globally, was detected using machine learning models. These models classified the hyperspectral reflectance profiles of the body surfaces of contaminated bees. The best-performing model, the linear discriminant analysis, achieved 98 % accuracy in discriminating contamination across species Apis mellifera, Melipona mondury, and Partamona helleri, with prediction speeds of 0.27 s. Our pioneering study introduced an effective method for discerning multiple classes of bees contaminated with pesticides using hyperspectral reflectance. An AI-driven spectral data analysis tool (https://github.com/bernardesrodrigoc/MACSS) was developed for the purpose of identifying and characterizing new samples through their spectral characteristics. This platform aids efforts to monitor and conserve bee populations and holds potential importance in environmental monitoring, agricultural research, and industrial quality control.

Farm to fork applications: how vibrational spectroscopy can be used along the whole value chain?

Vibrational spectroscopy is a nondestructive analysis technique that depends on the periodic variations in dipole moments and polarizabilities resulting from the molecular vibrations of molecules/atoms. These methods have important advantages over conventional analytical techniques, including (a) their simplicity in terms of implementation and operation, (b) their adaptability to on-line and on-farm applications, (c) making measurement in a few minutes, and (d) the absence of dangerous solvents throughout sample preparation or measurement. Food safety is a concept that requires the assurance that food is free from any physical, chemical, or biological hazards at all stages, from farm to fork. Continuous monitoring should be provided in order to guarantee the safety of the food. Regarding their advantages, vibrational spectroscopic methods, such as Fourier-transform infrared (FTIR), near-infrared (NIR), and Raman spectroscopy, are considered reliable and rapid techniques to track food safety- and food authenticity-related issues throughout the food chain. Furthermore, coupling spectral data with chemometric approaches also enables the discrimination of samples with different kinds of food safety-related hazards. This review deals with the recent application of vibrational spectroscopic techniques to monitor various hazards related to various foods, including crops, fruits, vegetables, milk, dairy products, meat, seafood, and poultry, throughout harvesting, transportation, processing, distribution, and storage.

Non-Destructive Classification of Organic and Conventional Hens' Eggs Using Near-Infrared Hyperspectral Imaging

Eggs that are produced using organic methods retail at higher prices than those produced using conventional methods, but they cannot be differentiated reliably using visual methods. Eggs can therefore be fraudulently mislabeled in order to increase their wholesale and retail prices. The objective of this research was therefore to test near-infrared hyperspectral imaging (NIR-HSI) to identify whether an egg has been produced using organic or conventional methods. A total of 210 organic and 210 conventional fresh eggs were individually scanned using NIR-HSI to obtain absorbance spectra for discrimination analysis. The physical properties of each egg were also measured non-destructively in order to analyze the performance of discrimination compared with those of the NIR-HSI spectral data. Principal component analysis (PCA) showed variation for PC1 and PC2 of 57% and 23% and 94% and 4% based on physical properties and the spectral data, respectively. The best results of the classification using NIR-HSI spectral data obtained an accuracy of 96.03% and an error rate of 3.97% via partial least squares-discriminant analysis (PLS-DA), indicating the possibility that NIR-HSI could be successfully used to rapidly, reliably, and non-destructively differentiate between eggs that had been produced using organic methods from eggs that had been produced using conventional methods.