Evaluación de la correlación entre las imágenes multiespectrales y el contenido de carotenoides en Jungia rugosa

Abstract

Carotenoids are compounds that have been extensively studied in different matrices due to their potential biological activity; being the antioxidant power the main property of interest. This brings the challenge of searching for new alternatives to facilitate their identification and quantification. In Ecuador, the use of multispectral images as a non-destructive technique to predict the presence of secondary metabolites in medicinal plants is limited. In this context, the objective of this study was to evaluate a possible correlation between reflectance values extracted from multispectral images with the concentration of total carotenoids in leaves of Jungia rugosa. In this study, total carotenoids were determined by spectrophotometry using two extraction methods. The first method was done by maceration of freeze-dried leaves and the second one from chloroform extracts obtained by percolation. Multispectral images were obtained using the Parrot Sequoia sensor at the collection sites Llaviuco, Mosquera, San José de Raranga, and at the University of Cuenca School of Chemical Sciences to evaluate possible variations in reflectance values at these locations. Multiple regression analysis reveals that both extraction methods are statistically significant (p<0.05) for predicting carotenoid content. However, the extraction method one (R 2 =0.28) and two (R 2 =0.36) show adjusted R² values that explain a low percentage of data variability. Although this research found a correlation between the RED Band (640-680 nm) and carotenoid content, it cannot be used as a model to predict carotenoid concentration. Therefore, these results should only be considered as a starting point for further analysis.