Assessment of Glucosinolates in Chinese Kale by Near-Infrared Spectroscopy

Using Near-Infrared Reflectance Spectroscopy (NIRS) to Predict Glucobrassicin Concentrations in Cabbage and Brussels Sprout Leaf Tissue

“…The models reported here outperform several others [27][28][29]. The model developed by Chen et al [30] performed better than the present study, and it was useful for qualitative determination of GBS in Chinese kale. The model by Chen et al [30] was built using a wider wavelength range and a greater sample size which could contribute to its better accuracy, although this is not completely clear as there are currently very few studies aimed at the development of GBS calibration models on vegetative tissue.…”

“…The model by Chen et al [30] was built using a wider wavelength range and a greater sample size which could contribute to its better accuracy, although this is not completely clear as there are currently very few studies aimed at the development of GBS calibration models on vegetative tissue. Additionally, the broad, diverse origin of a large number of cultivars that we obtained to develop this model may contribute to higher error compared to that seen by Chen et al [30]. Arguably, our models would more accurately re ect the diversity of materials breeders, producers, and processors would encounter in utilizing NIRS to characterize GBS concentrations for their particular needs.…”

“…Since humans eat leaf tissue of cabbage and Brussels sprouts rather than the seeds, having quanti cation methods for leaf tissue is more relevant. Six studies have attempted to develop quantitative calibration models for leaf tissue of several GSL-containing plants with various degrees of success [25][26][27][28][29][30], and four of these included GBS for calibration based on dry weight reference values (Table 1). Of those, Sahamishirazi et al [27] reported a model that had poor performance in broccoli orets.…”

“…This may explain the better performance of the GBS prediction model based on dry weight GBS concentrations. In comparison to the four other studies that developed calibration models for GBS in the typically consumed components of Brassica vegetables, we used a wider range of GBS concentrations (Table 3) than the other studies (Table 1) with the exception of Chen et al [30] who had a similar range but a slightly more right-skewed distribution. Model development.…”

Estimation of functional components of Chinese cabbage leaves grown in a plant factory using diffuse reflectance spectroscopy

“…When developing calibration models, relating the response variables and predictor variables, various methods, such as classical least squares, inverse least squares, principal component regression, partial least squares regression (PLSR), stepwise multiple linear regression (SMLR) and artificial neural networks, have been commonly used . In previous studies with Chinese kale and cabbage, the visible (VIS)‐near infrared (NIR) technique has been used to predict the content of individual glucosinolates and their total amount. It was revealed that, in terms of predetermination of the content of both individual and total glucosinolates, the NIR spectroscopy technique was feasible, reasonably inexpensive and highly accurate.…”