Rapid assessment of pork freshness using miniaturized NIR spectroscopy

The freshness of meat is an important quality attribute. Thiobarbituric Acid Reactive Substances (TBARS), a classical marker of lipid oxidation, is used as an indicator of freshness in meat. This study investigated the application of a miniaturized and portable Near Infrared (NIR) spectroscopy (1100–2200 nm) for rapid monitoring of TBARS in minced pork stored at 4 °C for 0, 2, 4 and 8 days. NIR data was pre-treated with standard normal variate (SNV), multiplicative scatter correction (MSC), and Savitzky Golay first derivative (FD). Models developed, based on full wavelengths, using partial least square regression (PLSR) showed good results in terms of their coefficients of determination (R2). The optimized result was obtained from MSC with R2 of 0.844 and RMSE of 0.099 mg MDA/kg meat in the prediction. To build a simpler model, optimal wavelengths were selected by successive projection algorithm (SPA) and weighted regression coefficients (RC) and used to develop four new models based on PLSR and MLR algorithms (SPA-PLSR, SPA-MLR, RC-PLSR, RC-MLR). The simplified model (RC-PLSR) revealed good results with R2p of 0.830 and RMSE of 0.068 mg MDA /kg meat. The promising result in this study indicated the potential of using a miniaturized NIR spectroscopy as a good handheld tool for rapid monitoring of TBARS values for assessment of freshness in minced pork.