Innovative and non-destructive technologies to evaluate quality of rocket leaves for ready to eat salads

‘Rocket’ is a collective name to indicate many species of green leaves belonging to Brassicacea family and are significantly consumed in the Mediterranean countries either as stand-alone salads or mixed with other vegetables. They are well known for their pungent smell, bitter flavor and high nutritional value. Rocket leaves are commercially grown as perennial and annual species, the former known as perennial wall rocket also known as wild rocket (Diplotaxis tenufolia(L.) DC.) and the latter named annual garden rocket (Eruca satvia Mill.). The Diplotaxis tenufolia plant can achieve a height of 80 centimeters (cm) and is characterized by a tap root and lengthy leaves. A typical leaf of Diplotaxis tenufolia is fleshy, oblong and deeply lobed, possessing sharp apexes. On the other hand, Eruca satvia Mill. grows to achieve a height of 40 cm possessing lyrate-pinnatifid leaves, having an enlarged terminal lobe and smaller lateral lobes with a rosette shaped arrangement of the leaves. Eruca satvia Mill. species as compared to the Diplotaxis tenufolia possesses a thin tap root and is characterized by a rigid unbranched stem. Both the species manifest similar morphological and nutritive aspects possessing a characteristic bitter taste based on the account of glucosinolates present. In the modern era, the consumer awareness regarding food safety, origins of the produce, nutritional value and demand for minimally processed fresh produce has led the food industry to explore rapid, reliable and cost effective methods for the evaluation of food products and their shelf-life since the conventional destructive analysis methods are time consuming, expensive, targeted and labor intensive. In this regard, non-destructive methods are gaining significant popularity which are assisting the food industry for the early fruits defect detection, fruits and vegetable classification on the basis of variety, maturity stage and origin and for the prediction of main internal constituents, mainly soluble solids and acids, and physical properties like firmness. On the industrial scale a significant weightage is given towards achieving fresh produce with superior quality in terms of vitamins, antioxidant activity, phenols and secondary metabolites. Rising concerns regarding the nutritional composition led many research works to evaluate the feasibility of the spectral profiles in the visible near infrared range (Vis-NIR), near infrared range (NIR) and hyperspectral images (HSI) for prediction and mapping of desired compounds in fresh produce. It is important to mention that non-destructive techniques cannot completely replace the conventional methods but can serve to assist these techniques saving time, expenses and labor. On the other hand, the non-destructive methods need no sample preparation once the model is developed making the prediction process quick. In this research work non-destructive techniques have been illuminated with respect to their potentiality in rocket leaves with special emphasis on hyperspectral imaging for the quality assessment of the fresh-cut rocket leaves accompanied by a basic introduction of the non-destructive image analysis techniques. In the first research work the feasibility of using spectral profiles for the estimation of the shelf life of the rocket leaves was evaluated using a multivariate accelerated shelf life testing (MASLT) approach. Spectral changes over time were modeled by using principal component analysis (PCA) and as variation to the conventional method, partial least squares (PLS) method. Kinetic charts were built fitting the first principle component (PC1) and the first latent variable (LV1) scores versus time. In both cases, the kinetics were described by a first order reaction, and the model performance was evaluated by the R2 values which ranged between 0.73 to 0.95 for samples stored at three different temperatures, one of them being the market temperature while the rest were categorized as accelerated temperatures which are usually higher than the market temperature. The cut-off value was calculated by judging the unacceptable spectra of samples at the accelerated temperature, as a result of which the shelf life of rocket leaves was estimated using the MASLT approach. The shelf life estimation was done using PCA based MASLT conventionally used as well as using a newly introduced methodology i.e. PLS based MASLT yielding encouraging results in both cases particularly in case of the PLS based MASLT. On the other hand, since the literature regarding quality evaluation of rocket leaves or any other leafy vegetables over time shows that the potential of hyperspectral imaging in the visible and near infrared regions has not been investigated pursing the aim of prediction and mapping of internal constituents. Hence hyperspectral imaging data was evaluated employing Partial Least Squares regression (PLSR) for the prediction of Vitamin C, ascorbic acid (AA), dehydroascorbic acid (DHAA), antioxidant activity and phenols in wild rocket (Diplotaxis tenuifolia) over a storage span of 12 days at 5oC. Hyperspectral images of the wild rocket leaves were acquired in the Vis-NIR (400-1000nm) and the NIR (900-1700nm) ranges using different data pretreatments and wavelength selection techniques. The model reliability was checked by the root mean square error (RMSE) and R2 values. Among the predicted parameters Vitamin C, AA, antioxidant activity and phenols were predicted satisfactorily in the NIR range. The prediction maps for the parameters were calculated to follow the changes over the storage period yielding more reliable results in the NIR range. All the results indicated that hyperspectral imaging combined with multivariate data possess the capability to provide reliable information regarding the shelf life estimation of the rocket leaves as well as for the prediction and mapping of the internal constituents.