Determination of aflatoxin B₁ level in rice (Oryza sativa L.) through near-infrared spectroscopy and an improved simulated annealing variable selection method

Direct quantification analysis of near-infrared (NIR) spectra is challenging because the number of spectral variables is usually considerably higher than the number of samples. To mitigate the so-called curse of dimensionality, variable selection is often performed before multivariate calibration. There has been much work in this regard, where the developed variable selection method can be categorized as individual variable selection, such as uninformative variable elimination or variable importance in projection, and continuous interval variable selection method such as interval partial least squares or moving window partial least squares. In this study, a new individual variable selection method, modified simulated annealing (MSA), was proposed and used in conjunction with the partial least squares regression (PLSR) model. The interpretability of the selected variables in the determination of aflatoxin B₁ levels in white rice was assessed. The results revealed that the PLSR model combined with MSA not only yielded higher accuracy than the full-spectrum PLSR but also successfully shrank the variable space. The developed simplified PLSR model using MSA produced satisfactory performances, with root mean square error of calibration (RMSE_C) of 0.11 μg/kg and 0.56 μg/kg, and root mean square error of prediction (RMSE_P) of 7.16 μg/kg and 14.42 μg/kg, were obtained for the low-aflatoxin B₁-level- and high-aflatoxin-B₁-level samples, respectively. Specifically, the MSA-based models yielded improvements of 97.80% (calibration set) and 44.62% (prediction set) as well as 95.85% (calibration set) and 62.57% (prediction set) for both datasets when compared with the full-spectrum PLSR (low aflatoxin: RMSE_C = 5.02 μg/kg, RMSE_P = 12.93 μg/kg; high aflatoxin: RMSE_C = 13.50 μg/kg, RMSE_P = 38.53 μg/kg). Compared with the baseline method of simulated annealing (SA) (low aflatoxin: RMSE_C = 0.21 μg/kg, RMSE_P = 9.78 μg/kg; high aflatoxin: RMSE_C = 12.27 μg/kg, RMSE_P = 38.53 μg/kg), the MSA significantly improved the predictive performance of the regression models, with the number of selected variables being almost half of that in the SA. A comparison with other commonly used variable selection methods of selectivity ratio (low aflatoxin: RMSE_C = 6.09 μg/kg, RMSE_P = 13.75 μg/kg; high aflatoxin: RMSE_C = 13.74 μg/kg, RMSE_P = 41.13 μg/kg), uninformative variable elimination (low aflatoxin: RMSE_C = 0.32 μg/kg, RMSE_P = 5.11 μg/kg; high aflatoxin: RMSE_C = 3.80 μg/kg, RMSE_P = 17.76 μg/kg), and variable importance in projection (low aflatoxin: RMSE_C = 2.67 μg/kg, RMSE_P = 10.71 μg/kg; high aflatoxin: RMSE_C = 13.51 μg/kg, RMSE_P = 32.53 μg/kg) also indicated the promising efficacy of the proposed MSA.