Qualitative differentiation of non-deteriorated and deteriorated codonopsis radix using a portable spectrometric system with a multispectral sensor chipset

Abstract

Codonopsis Radix is a commonly used Chinese medicine and is also a food-medicine homology plant used as a food ingredient in China and Southeast Asia. Its nutritional profile, rich in polysaccharides, triterpenoids, polyacetylenes, and essential oils, renders it prone to deterioration during storage. In this study, we developed a portable, streamlined spectrometric system applicable for the rapid identification of whether Codonopsis Radix samples are deteriorated. This system centers around a multispectral sensor chipset (AS7265x) equipped with an 18-channel array spanning the visible to shortwave near-infrared spectrum (410–940 nm). We validated the system's efficacy through chemometric methods, including Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA), and Orthogonal Projections to Latent Structures Discriminant Analysis (OPLS-DA). Additionally, we compared its performance to that of a commercial dispersive spectrometer (USB2000+). The validity and robustness of the OPLS-DA models built from the AS7265x and USB2000+ calibration spectral data were demonstrated based on the cross-validation and permutation test results, which indicated that the spectral data from the AS7265x device were comparable to those from the USB2000+ device in building a robust model that could effectively discriminate between non-deteriorated and deteriorated Codonopsis Radix samples. The prediction accuracies of the OPLS-DA methods based on the spectral data from the AS7265x and USB2000+ devices were both 95% under external validation. The experimental results confirmed that the proposed system is a powerful tool for practical, cost-effective applications in monitoring and assessment of Codonopsis Radix quality from major production areas in China.