[Effects of different stir-frying durations on non-volatile metabolites and sensory quality of stir-fried Paeoniae Radix Alba utilizing non-targeted metabolomics and intelligent sensing technologies].

To address the unclear patterns of change in non-volatile metabolites and sensory quality characteristics during the stir-frying of Paeoniae Radix Alba, non-targeted metabolomics, electronic eye, and electronic tongue technologies were employed to comprehensively analyze the component, color, and taste of stir-fried Paeoniae Radix Alba under different stir-frying durations. UHPLC-Q-Exactive Orbitrap MS was utilized to reveal the dynamic changes in metabolites in Paeoniae Radix Alba across different stir-frying durations, and a total of 59 non-volatile metabolites were identified. Among them, 11 key differential non-volatile metabolites with VIP≥1 and P<0.05 were screened. Changes in four colorimetric values and eight taste values during the processing of Paeoniae Radix Alba were explored using the electronic eye and electronic tongue, respectively. Integrated analysis of the component, color, and taste reveals a strong correlation between the changes in color and taste during the processing of Paeoniae Radix Alba and the changes in metabolites, such as the initial increase and subsequent decrease in oxidized paeoniflorin content and the increase in catechin content. Meanwhile, based on data fusion, a support vector machine(SVM) model was developed to differentiate Paeoniae Radix Alba stir-fried for different durations, significantly improving classification performance and achieving an accuracy of 100%. This study reveals the influence of stir-frying on the metabolites and sensory properties of Paeoniae Radix Alba and establishes a rapid method for evaluating the processing degree of stir-fried Paeoniae Radix Alba, providing important guidance for quality control and intelligent production of stir-fried Paeoniae Radix Alba decoction pieces.