High-dimensional colorimetric sensor array based on composite dyes for rapid detection of antioxidant properties...

Goji berry juice (GBJ) is a highly popular healthy drink among consumers. Rapid monitoring of the antioxidant properties of GBJ is crucial and necessary. A colorimetric sensor array (CSA) based on composite dyes was used for the rapid detection of the antioxidant properties of GBJ. High-dimensional spectral fingerprint information combined with chemometric methods was used to construct a quantitative prediction model. The effects of different preprocessing and feature screening methods on model prediction were compared. Machine learning algorithms were used to establish quantitative prediction models. The CARS-PLSR model achieved accurate quantification of total phenol, with Rp and RPD values of 0.95 and 3.18, respectively. For DPPH free radical scavenging rate, the CARS-LSSVR model had an Rp value of 0.95, an RPD value of 3.33, and an RMSEP of 0.04. The proposed method in the research provides a new idea for the rapid quality monitoring of the fermentation process of GBJ.