Predicting water-holding capacity of intact chicken broiler breast fillets with Vis/NIR spectroscopy

Abstract. The ability of using visible and near-infrared (Vis/NIR) spectroscopy to predict water-holding capacity (WHC) of intact chicken broiler breast fillets (pectoralis major) was assessed in this study. Boneless and skinless chicken fillets (214 in total) were procured from a commercial processing plant and spectra were collected on the skin side of fillets in the whole Vis/NIR spectral range (400-2500 nm). Prediction models for three different WHC indicators (Drip loss, DL; Water gain, WG; Expressible fluid, EF) using Vis/NIR spectroscopy were established through partial least square regression (PLSR) combined with mathematical pre-treatments, such as 1st or 2nd order derivative, standard normal variate (SNV) and multiplicative scatter correction (MSC). Predictive ability of models were assessed by correction coefficient of validation (R_v), root mean square error of validation (RMSEV), ratio performance deviation (RPD) and range error ratio (RER). A medium predictive ability was obtained for DL (R_v=0.78; RPD=1.60; RER=6.98), WG (R_v=0.76; RPD=1.54; RER=6.09) and EF (R_v=0.69; RPD=1.38; RER=6.29). Conclusively, results suggest that Vis/NIR spectroscopy is a useful tool for predicting WHC traits of intact broiler breast fillets; however, the predictive ability is moderately influenced by the reference methods selection.