Predicting soluble solid content in intact jaboticaba [Myrciaria jaboticaba (Vell.) O. Berg] fruit using near-infrared spectroscopy and chemometrics

• Published in: Food chemistry Vol. 159; pp. 458 - 462
• Main Authors: Mariani, Nathália Cristina Torres, da Costa, Rosangela Câmara, de Lima, Kássio Michell Gomes, Nardini, Viviani, Cunha Júnior, Luís Carlos, Teixeira, Gustavo Henrique de Almeida
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 15-09-2014; Elsevier
• Subjects: Algorithms; Biological and medical sciences; BP-ANN; Calibration; Food toxicology; Fruit - chemistry; Least-Squares Analysis; LS-SVM; Medical sciences; Multivariate Analysis; Myrtaceae - chemistry; Neural Networks (Computer); NIR spectroscopy; Plant Extracts - chemistry; PLS; Reproducibility of Results; Spectroscopy, Near-Infrared; Support Vector Machine; Toxicology; Variables selection; LS-SVM; Variables selection; NIR spectroscopy; PLS; BP-ANN; Fruit; Selection; Content; Plant origin; Prediction
• ISSN: 0308-8146; 1873-7072
• DOI: 10.1016/j.foodchem.2014.03.066

•NIRS was applied to determine soluble solid content (SSC) of intact jaboticaba fruit.•PLS, iPLS, GA, SPA, BP-ANN, and LS-SVM were applied to building calibration models.•The PLS model produced prediction accuracy (R2=0.71, RMSEP=1.33°Brix, and RPD=1.65).
The aim of this study was to evaluate the potential of near-infrared reflectance spectroscopy (NIR) as a rapid and non-destructive method to determine soluble solid content (SSC) in intact jaboticaba [Myrciaria jaboticaba (Vell.) O. Berg] fruit. Multivariate calibration techniques were compared with pre-processed data and variable selection algorithms, such as partial least squares (PLS), interval partial least squares (iPLS), a genetic algorithm (GA), a successive projections algorithm (SPA) and nonlinear techniques (BP-ANN, back propagation of artificial neural networks; LS-SVM, least squares support vector machine) were applied to building the calibration models. The PLS model produced prediction accuracy (R2=0.71, RMSEP=1.33°Brix, and RPD=1.65) while the BP-ANN model (R2=0.68, RMSEM=1.20°Brix, and RPD=1.83) and LS-SVM models achieved lower performance metrics (R2=0.44, RMSEP=1.89°Brix, and RPD=1.16). This study was the first attempt to use NIR spectroscopy as a non-destructive method to determine SSC jaboticaba fruit.