Accuracy of near infrared spectroscopy for prediction of chemical composition, salt content and free amino acids in dry-cured ham.

The capability of near infrared (NIR) spectroscopy was examined for the purposes of quality control of the traditional Slovenian dry-cured ham "Kraski prsut." Predictive models were developed for moisture, salt, protein, non-protein nitrogen, intramuscular fat and free amino acids in biceps femoris muscle (n = 135). The models' quality was assessed using statistical parameters: coefficient of determination (R(2)) and standard error (se) of cross-validation (CV) and external validation (EV). Residual predictive deviation (RPD) was also assessed. Best results were obtained for salt content and salt percentage in moisture/dry matter (R(CV)(2)>0.90, RPD>3.0), it was satisfactory for moisture, non-protein nitrogen, intramuscular fat and total free amino acids (R(CV)(2) = 0.75-0.90, RPD = 2.0-3.0), while not so for protein content and proteolysis index (R(CV)(2) = 0.65-0.75, RPD<2.0). Calibrations for individual free amino acids yielded R(CV)(2) from 0.40 to 0.90 and RPD from 1.3 to 2.9. Additional external validation of models on independent samples yielded comparable results. Based on the results, NIR spectroscopy can replace chemical methods in quality control of dry-cured ham.

An attempt to predict pork drip loss from pH and colour measurements or near infrared spectra using artificial neural networks.

The ability to predict meat drip loss by using either near infrared spectra (SPECTRA) or different meat quality (MQ) measurements, such as pH(24), Minolta L(∗), a(∗), b(∗), along with different chemometric approach, was investigated. Back propagation (BP) and counter propagation (CP) artificial neural networks (ANN) were used and compared to PLS (partial least squares) regression. Prediction models were created either by using MQ measurements or by using NIR spectral data as independent predictive variables. The analysis consisted of 312 samples of longissimus dorsi muscle. Data were split into training and test set using 2D Kohonen map. The error of drip loss prediction was similar for ANN (2.2-2.6%) and PLS models (2.2-2.5%) and it was higher for SPECTRA (2.5-2.6%) than for MQ (2.2-2.3%) based models. Nevertheless, the SPECTRA based models gave reasonable prediction errors and due to their simplicity of data acquisition represent an acceptable alternative to classical meat quality based models.